Jon Gould is the founder of Foxley Talent, a recruiting agency focused on Python and Django developers. We discuss what he's seeing in the market these days, the benefits of attending Python & Django conferences, and why humans matter more than ever in our new AI age.
🔗 Links
📚 Books
📦 Projects
🎥 YouTube
Will (00:01)
Hi, welcome to another episode of Django Chat. I'm Will Vincent with Carlton Gibson. Hey Carlton.
Carlton (00:07)
Hello well.
Will (00:08)
And we're very pleased to have John Gould from Foxy Talent back on after three years. Welcome, John.
Jon (00:14)
Yeah, thanks very much guys. Nice to be back on overview.
Will (00:19)
So you and I have seen a bunch of each other this year because at the various conferences, maybe I'll start with the softball. where have you traveled this year? Because you've been everywhere.
Jon (00:28)
Yeah, I've been to the two big Django conferences, so DjangoCon Europe in Athens, Greece, and then in DjangoCon US in Chicago earlier in August. And then I went to PyCon US over Long Beach as well. Those were my sort of three big events. yeah, I've seen one or either of every single one of them.
Will (00:53)
That's right. That's right. Yeah, two for me, one for Carlton.
Jon (00:56)
Yeah.
Will (00:58)
well so maybe and o sorry, for introduction, for those who don't know, you're at Foxy Talent, like the preeminent Django recruiting agency. what should I add? What's the pitch these days?
Jon (01:09)
Yeah,
suppose like, yeah, Foxy Talent has just turned five years old. I've recruited Python, Django Space and developers since 2008. So coming up to 20 years doing that. yeah, as I suppose, a career recruiter, I'm also an individual member of the Django Software Foundation and Foxy Talent is one of the corporate sponsors. So in the five years I've been going, I've donated over 20,000.
dollars towards that. I'm a regular at conferences, yeah, all everywhere. If you go on LinkedIn, won't be able to spend it, scroll for a couple of minutes without seeing me somewhere on the floor.
Will (01:55)
Well so it's
Carlton (01:53)
Yeah.
Will (01:58)
I I was just gonna say, so the conferences, so as a as a programmer, right, people go there to give talks, to network, to maybe look for jobs, to find out what's happening in the industry. What are you know, as a recruiter, what are you looking for? Is it similar? Are there differences? Like what do you pick up when you go to these events?
Jon (02:13)
Yeah, suppose the probably at DjangoCon US is very much for me it's the that being in the States is where most of my business now is. I probably in the last 12 months I've been 70 % of my placements and revenue has come from within the United States. So being there, it's being in the room with all the right people. Whether yeah, whether there's talks happening, but yeah, for me it's
it always started out being I just want to be around the people who have given up their own time, the best, you know, who I would see as like the best people in the room. Now it's changed a little bit with like fake candidates and all of this. I'm now in a room with just real Django people.
Carlton (02:57)
If they're at the conference they must exist.
Jon (03:00)
exactly, yeah, yeah, there is proof of life that everyone's real and it's not an AI filter. But it's, also these events are so important for the community. If you can...
help out by sponsoring and for me, having a stand where people can talk to me as well is a byproduct of that. It's worth doing and I like to help where I can. Django Day Copenhagen was a week ago, a week or 10 days ago, and
Will (03:31)
Yeah.
Jon (03:32)
I couldn't go but somebody I met at DjangoCon Europe really wanted to go. They...
had helped me out with some stuff on my website so I bought them a ticket and did like the supporter bit on top to help the guys run that conference. just trying to help those little community events happen where they do.
Will (03:53)
Yeah, and I I should
Carlton (03:54)
And you
Will (03:54)
just I'll say one more thing, Carlton, and then you can go. I was just gonna say, I remember having this conversation with multiple people at DjangoCon US saying, do you know John? If you don't, you should go talk to him because no matter how happy you are with your job, like you should know what's out there, right? Like you you need to
Jon (04:08)
Yeah.
Will (04:09)
get information about what the market's saying, how your employer's doing. And then I know for you, like you have these long term relationships. So even if someone's happy now, you build you know, you get the contacts, you understand their situation, and then two, three, four years, you know
this this position came up and this person I've met and talked to at conferences would be a good fit. So people shouldn't be shy about talking to you. which most people aren't, but a couple of times they need the nudge of like, no, it's like you should talk to John. Yeah.
Jon (04:36)
It's Yeah,
think that again is another key reason I go is to meet people. I mostly speak to people in my day job when they need something, when it's a transaction, when it's I'm looking
Will (04:50)
Mm. Yeah.
Jon (04:51)
for a role or you've got a job that I want to find out more about. It's a transaction a lot of the time. But by going to conferences, you meet people when just, it's not organic because you're at a conference, but you meet people.
and you can have a conversation and build a relationship and get to know somebody when there is nothing on the table. It's just, yeah, it's nice. then I find
Carlton (05:15)
Yeah.
Jon (05:17)
when, if a job comes in two years later that they said, I'd love to work in X industry. I will come back to them and go, hey, look, I know you may not be looking, but we spoke two years ago at this conference. I've got something that might interest you. It's a lot.
easier like that and it's the enjoyable part of the job. I much prefer to speak to people when they are happy in work and not looking than I've been laid off and I need to get something before the money runs out.
Carlton (05:49)
Yeah, no, 'cause then it's desperate, right? It's like
Jon (05:51)
Yeah. Yeah. And I mean, over the course of a year, I'll probably make between 15 and 20 placements. So I'm 15, 20 calls a week
with people.
Carlton (06:03)
Yeah.
Jon (06:03)
It's such a small proportion I end up actually putting into a job because that's where the, yeah, the nature of it.
Carlton (06:14)
Yeah. I think when we chatted before the the the business was changing a bit, but like it must have changed quite a lot over the last two, three years.
Jon (06:25)
Yeah, well, I built Foxley to, I had a couple of guys and at one point there was four of us working in an office in the town near where we are and it just, it was through, it grew through like a desire to, and maybe old habits of building something and diversifying and creating additional revenue streams and that sort of thing.
the number at the top never really changed. So whether it was me on my own or me and two or three others, the placements
Carlton (07:01)
Yeah.
Jon (07:02)
never really changed. So I had to make difficult decisions to scale back and for the last, well over 12 months now I've been on my own again and it's been probably my most fulfilling year. I've been close to everything. I've been able to lead my own projects as well and like.
I actually quite enjoy travelling on my own to these things as well. Like, going for an airport
Carlton (07:27)
Yeah, right.
Jon (07:27)
with no stress, don't have to worry about anyone else, like, I can get there. It's, yeah, those sort of things work quite well. No
Carlton (07:32)
So you haven't got the kids in tow.
Jon (07:34)
kids, no wife, no colleagues, it's perfect.
Carlton (07:37)
Yeah
right, okay,
Will (07:38)
Yeah.
Carlton (07:39)
fair enough. So you've started a new podcast, Sean, that's why we really wanted to get you on 'cause you know, tell us tell us about the new the new thing. What's it go on, give us the rando.
Jon (07:48)
Yeah, yeah, I have. So
I've started a podcast called People or Code. There's now three episodes, which includes my story to start it off, which are out. And I've got another three recorded, which are going out over the next couple of Mondays. And it's basically this conversation around the point in an engineer's career where the decision is to go
further in their career into people management or into code. And I suppose from the five conversations I've had, also now like what impact is AI agents and what impact is that going to have on the path that people choose? And so the sort of the goal behind this is to start that conversation so that people who have stood at the fork in the road, the crossroads have somewhere to go for a bit of guidance and advice, but also to help the people who are building teams in.
26, 27 onwards, learn and sort of make those good decisions as well about how they should shape a team.
Carlton (08:58)
I mean for me that's the classic crux point because you know, when I was coming up I never wanted to be a manager, I wanted to be a coder forever. But if I said, No, you've got to move into management, you've got to move into management, you've got to move into management. And I remember seeing some article about how, you know, it but get out of coding before you're forty. And I'd just turned forty and I was like, Well, a bit late now.
Jon (09:22)
Yeah, but it was the same in the recruitment world. I yeah, enjoyed doing the job of a recruiter and I got to a point where I had a couple of people around me and I was then moved in. I had opportunities to move up and everybody was like, move up, get off the tools, like get a job before your head goes gray and you're knocking on, yeah, you're having to go to conferences and stuff. And it was...
Yeah, it was not forced upon me, I enjoyed it. I enjoyed every minute of building a team out and like the camaraderie that came with that. And some of the guys who were in my team are still really good mates. Like Will, you came over to the UK, we played golf with one of the guys who used to work for me, who's running his own recruitment business as well. It was brilliant, but I always missed the job, the job that I loved, which was recruiting. And even when I...
started the business again on my own at Foxley, went old habits of building a team, fell back and I got further away from doing the job that I love. So I'm now facilitating that conversation for engineers and like, yeah, because there is no right answer. I think that the good thing about the length of a career is you can always go back and unpick it and you can go back to that.
with experience under your belt and try either route at any time. And some of the guests I've had on have done both and moved, yeah, moved, done either route and loving it or have done one route and changed. So yeah, some great stories that are coming out. So shame this plug, if anyone's,
Carlton (11:05)
Super, I think that's a
Jon (11:07)
yeah, if anyone's done that and wants to come on and talk about
Will (11:10)
Mm-hmm.
Jon (11:11)
their story, then yeah, we're pleased to get in touch.
Will (11:15)
Yeah, I mean
Carlton (11:15)
I think it's a great topic.
Will (11:17)
No, it's it it's it's sort of the topic. I mean, Carlton, I wonder, not to psychoanalyze you, but you know, you you did your PhD in philosophy, like most people most people like if they start coding, you know, in their early mid twenties, they join a reasonably sized company, you get a lot of pressure four or five years in. Like if you're not managing, like there's not a role for you. I wonder, like I mean, you know yourself well, but also you were consulting and stuff, right? Because it's pretty hard to avoid that in a big company like, you know, after
four or five years because the company wants to scale up, right? And so you're more you're you're more valuable as a manager.
Carlton (11:53)
Yeah, no, I mean and I I
when I started two two six, or after the dot com bubble and I'd been in I'd been in the kind of nice shelter of academia with a you know a nice grant to to fund me during the the the bust after the dot com bubble. So I'd seen it I'd seen it happening. I was like, that doesn't look good. And I I always had this idea about the one the one man business model, keep it small, keep it just me, keep because as soon as you've got as soon as you've got staff, then you've got a commitment because th those people's livelihood depend upon you making payroll every month.
And I remember when I was beginning I worked did some freelance work for an agency that were winning all the prizes. They were a iOS agency and they they built Facebook's app and whatnot and they had big offices in Barcelona, 50 people, they were winning awards left, right, and center, with and within about th two or three months they just disappeared because for for whatever reason the pipe the sales pipeline dried up and then they had this massive payroll which they couldn't afford. So everybody got laid off and the the the sort of core team got aqua high by Facebook.
Facebook and they went and built the the app in-house. And it was like, no, that's why you don't build a massive business with a massive payroll, w which you can't, you know, see
see the the pathway clear. I th th there was this Bill Gates line about always he w always wanted Microsoft to have enough cash on hand to go a year without making a single sale. And I kinda took that very much to hand to heart, but like at my scale, it's like you must have enough reserve so that, you know, if push comes to shove, there's a window where you can find a job without being in the panic situation that you were talking about before, John. but yeah, everything you're talking about here and
maintaining the position of an individual contributor, that speaks right to my heart because it's what I've tried to do my entire career. I'm not interested in
Will (13:39)
Ha ha ha.
Carlton (13:40)
being a manager, which we'll you know, we'll come to a different topic shortly, I guess.
Will (13:44)
Yeah. But
I I would say you gotta be pretty good to get away with an IC career. you know, I mean if you're I mean, no one self assesses as a mediocre programmer, but you know, it is true that some people are great managers and maybe mediocre programmers. I mean that's right. And not to turn this into one of your episodes, John, but
Carlton (13:59)
Yeah.
Will (14:00)
like their
Jon (14:00)
Hmm.
Will (14:01)
their different skills is sort of the core thing and yet they're thrust upon you. so
Jon (14:07)
Yeah,
yeah, and I think as well, like with an open source, cards are working as a fellow, as an individual contributor. Like that is a role, I know not often open to many, but it is in software and especially in open source, like giving back and using your skills in that way is also a, is an option as a route or as a goal that people may have is like, it's not managing people, but it is doing more than just.
Yeah, being a principal engineer somewhere. Yeah.
Will (14:35)
I mean it's worse. It's it's manj it's
managing the community. But yeah, that's a that's a separate topic. I mean,
Jon (14:40)
Yeah. Yeah.
Will (14:42)
I would just say my l my
Carlton (14:43)
I love it.
Will (14:44)
little plug is is like I I've managed people and I think for most people, I think it's good to do it and try it, but recognize it doesn't have to be forever. Because so for me now, like I'm managed by 30 somethings, right? 30 somethings run the world in tech, right? And so
When
there's things I disagree with, like I have a lot more empathy for the situation they're in because I was in some version of it before, right? Whereas if I don't if I'd never been in those shoes, I might be a little more like, you know, lacking an understanding for the situation. so anyways, yeah, people should listen to the podcast and it's a perennial topic.
Jon (15:21)
Yeah, forgive the rustic nature of it, I'm still figuring
all that bit out, but...
Carlton (15:26)
You mate, it took us
Will (15:26)
no, yeah.
Carlton (15:27)
years. We're two hundred episodes in and still learning.
Jon (15:29)
Yeah
Will (15:31)
Well, I wanted to ask you about so we
Jon (15:32)
Heh.
Will (15:32)
mentioned so Carlton that you said the the aqua hire of the agency. I think this came up, John
Carlton (15:36)
Yeah.
Will (15:37)
John, in in the US. Like, I won't name names, but there's a number of high profile Django people who are now at Frontier labs or doing, you know, AI stuff. Do you see how much do you see that happening, right? Because AI salaries are a bit better than slinging Django code. Is that is that true? Or what are you seeing out there in the market?
Jon (15:59)
I think there's, if you look at where the investment funds are putting their money, it's anything which is AI heavy at the moment, and therefore they are gonna have more, more carrot, they might have, yeah, they are throwing money at their staff because they've got to, they wanna try and attract the best, they're in a bit of an arms race for people, and you have to get their product out first, but then you've also, that,
there's a lot of fatigue in the job market where people have been hired by a startup or a scale up and have been let go a year, 18 months later. So I think there's, although yeah, you can chase something shiny and take a role with an AI company. I think there's a little bit of trepidation around it because you don't wanna be, yeah, what if staying put in your...
slinging Django code role is actually the safer long-term play.
Carlton (17:00)
But it might might still be there in a year or two.
Jon (17:03)
Yeah,
yeah, is 21 years old. There's a whole ecosystem around it and AI is, I don't know, it's in the job sector and hiring, people are using it, it's changing the way people are writing code and are evaluating engineers. It's having a knock on to junior roles, definitely, but I think there's yeah, still a...
Good job market for Python Django engineers.
Carlton (17:35)
So just nip into that junior question at the moment. I mean it looks like a serious crisis, sustainability crisis, in that if we're if n if we're not hiring juniors and we're not training them up, we're not gonna have seniors in half a dozen years time. does how how does that look from your end? Is that a real concern? Is it are folks still
Jon (17:57)
Yeah.
Carlton (17:58)
training folks? Or
Jon (18:03)
Yeah, there are juniors coming to the market. think grads coming through from universities and colleges who've got links into employment, there's still grand paths into bigger organizations. It's more like those boot camp and reskillers who are career switches who are probably not getting much.
anywhere near as much opportunity as before. don't know of anywhere near as many companies that are actively hiring juniors. There's a handful that you sort of see every now and then, but they all have their, A, work cut out identifying which person's gonna be the right one or two to bring in, but also, you know, their...
they probably got their formula, like they might have a link to a college university locally who are doing it. And I think partly this is because of the AI tools being the junior developer, but also remote work. think remote work is possibly detrimental to juniors integrating and learning, like the nuance of sitting in a row of desks where somebody's working on.
on something and they can just lean over and go, hey, can you look at this? Can you explain how this works? But yeah, if you look at the sprints and things at Django cons or, know, there are people of every level learning from each other just by osmosis
Carlton (19:35)
Yeah.
Jon (19:36)
of being in the same room. And I think for remote working that has had a real impact on the, and juniors can learn and can progress by doing it that way. But I think it's just a negative.
aspects of remote working is the things that people haven't hired juniors because of the speed to get up and running. And they may as well wait six months and hire somebody with more experience or wait and increase budget and hire a mid or a senior.
Carlton (20:09)
Yeah, right, okay. Yeah, that analogy to the
Will (20:11)
I mean, but that was always
Carlton (20:12)
sprints is close is a good one 'cause it is exactly like that. The the the the the high bandwidth of bin in the room just makes all the difference. Will, go on.
Will (20:21)
I was gonna say, I mean, it's not I mean it's always been the case, you know, the sweet spot is someone two, three, four years into a career where they've like gotten over the you know someone's trained them but then they kinda know something but they're still malleable. You know, whereas you bring Carlton or me into an organization, we're gonna have opinions. You get you're you know, it's it's a it's a
Carlton (20:38)
You get what you get with
Will (20:42)
it's a solid piece of of wood, you know. So, you know, but I I mean on the hiring front I
Everything
you said just resonated. Like the big companies are still I mean, people don't understand that like Google, Facebook, like the average age is young. Like it's like twenty seven, twenty six, twenty seven, like and you know, they're hiring less, but they're still like my brother works at Columbia University in recruit career services, right?
Like they're they're their grads get jobs, right? It's no problem, right? Because they have the pipeline. So like the you know, and these big companies are investing more than ever in offices, right? Like Manhattan is like a lot of software now. so you still they still want smart young people. The problem is, as you said, like maybe a boot camp person or someone who hasn't gotten that on the ground training, it is much harder. Like I mean, consultancies I've spoken to and you've probably spoken to, John.
it's harder for them to hire a you know, a grad from a from a boot camp, right? Whereas before they could have, now they can kind of swap that out for an agent. So, anyway, so two things are true. I'm just agreeing with you. The big companies still need people.
Jon (21:52)
Yeah.
Will (21:52)
A lot of those people are more junior, but I think that mid level or mid career is really getting nailed because yeah, like why train a person when you can train an agent and do a harness? Unfortunately a lot of people have that mindset.
Jon (22:07)
I talked to juniors about engineering their job search, so almost like reverse engineering. So if you're based in Chicago and you're a junior engineer, you're looking for remote jobs against junior engineers everywhere in the country, your unique selling point is
is your experience. You haven't really got any. So actually, if you narrow down your search and using tools like LinkedIn, you can do like a search for people by typing in, you know, Python, Django, junior developer into the search that and you can.
search for people who have got the job that you want. So you can look for people who are junior developer and you can look by location in Chicago. And in theory, you should be able to find the companies that have already got junior developers working for them. Knowing that people don't update their LinkedIn as often as they want, they've probably been promoted and haven't been back in to change the job title. But that company specifically is hired a junior in the last few years. So
That's a company to approach and say, hey, I'm a junior, I'm looking for work. Doing it that way around rather than searching for jobs. It works. doesn't work every time, you've got, yeah, it's by being in your USP is now that you can get to the office or you're in the same location as opposed to you've you've written a bunch of code.
Carlton (23:44)
So I always thought that the o return to offices was like the revenge of the companies, but it might actually be revenge of the junior developer.
Jon (23:52)
Yeah. Yeah.
Will (23:52)
I think it's just a way to get rid of
old people without violating firing laws, Carlton. Well no, because
Carlton (23:57)
Well, if you make make old people come to the office they quit.
Will (24:01)
no, because yeah, if someone who's like in their thirties, forties with kids who, you know, you say, I'm not firing you 'cause of the age or this, that and the other thing, it's like y you know, you gotta be in the office five days a week. I mean I I don't wanna get too conspiratorial, but I'm very confident that a lot of it is that.
Carlton (24:20)
Okay, interesting.
Will (24:20)
I don't know, John, what
do you think? I'm just shooting my mouth off, but I've I've heard that a lot. Okay, okay. Well I did I
Jon (24:22)
I'm not getting involved. Yeah.
Carlton (24:26)
Very diplomatic.
Will (24:28)
on the on the hiring front, I mean one thing that must be hard, I haven't done a I haven't hired someone in a while, but it it used to be the case that I would always and I'd tell people like put some projects on GitHub or, you know, Codeberg, you know, that you would write and then for the interview we could talk about what you built rather than whiteboarding problems.
But now that's gone because everything is AI, right? Even if you could do it manually, basically everyone's doing something AI. And if they're not like there's no way to check that they ought that they didn't. So I wonder what you're s if you're s what are you hearing on that, John? Because I I find that's a real loss. I used to you know, I guess you can quickly find out if you talk to someone, you can say, walk me through this and they're like, here's the prompts I used. But it used to be
Jon (25:13)
Yeah.
Will (25:13)
that as a screen, if someone had a number of projects that were open source,
You know, I could evaluate them, whereas now I kinda have to talk to to see if they know what they're talking about. I don't know, what do you find?
Jon (25:25)
I
don't think talking to someone during an interview process is a bad thing. think
Will (25:31)
Yeah.
Jon (25:31)
it's that there's too
Carlton (25:35)
Right, but you it's how you get past the first big pile, right?
Jon (25:40)
much of hiring is now automated. It's in a world where AI is being used a lot. Hiring should be more human.
the way that even from a point of when you're writing your CV, your resume, avoid the instinct, the going in or optimising it for every role because you'll just come out sounding the same as everybody else. And I think if you've been searching for a while, you'll be tempted, I've got to change something, I've got to optimise it for this role. It looks...
Any document that has been pushed through an agent will look the same and sound the same. yeah, even from that point, then the people reviewing them, like smaller companies don't tend to have, they tend to have a person reviewing or an external recruiter reviewing resumes and CVs rather than an applicant tracking system where it would just search just the keywords. So again, be more human leaning to like the...
the reason why you're looking and your backstory, how you got into coding in the first place, what you're still passionate about. And then the interviews, I can't stand things like the leak code processes, any interviews where you're being set up a task,
Carlton (27:07)
yeah, yeah, yeah.
Jon (27:09)
this just fictional, just review some code or take a pull request.
Carlton (27:13)
Yeah.
Jon (27:14)
No one's too busy just to go, hey, have a look at that, solve that for us. And if you do,
If you work where you're using prompts in your coding and generating code that way, well then run your interview process like that and say we want you to use whatever tool you're most comfortable with or we use Claw Code or whatever. And then you review it together like you would in the workplace. That's the biggest thing is that point of testing somebody.
If it's miles away from the way that you work, what's the point? If you're to send somebody off to do something and go, hey, please don't use it. We want to see actually how you code. Then you'll filter people out that way because you'll know straight away if they follow the instruction or not. it is, real. think just hiring needs to be more human in general. Not just to keep me in a job, but it's...
you're hiring a person, keep it on in conversation.
Carlton (28:27)
I as you describe that pro, no, just as you
Will (28:27)
I have more questions, but Carlton on yeah, I don't wanna
Carlton (28:30)
describe that process, I'm like I'm thinking to the to the times that I've been responsible for f filling a role and you put a you know, you put a thing up and you get f and this was before the days of AI, right? It i it you'd get far too many applicants and it was far too difficult to to get through them and be you know, and then far too difficult to do the the the the interview process fairly and and and and at
reasonable personal expense, like you know, reasonable
Jon (28:59)
Yeah.
Carlton (29:00)
time investment and what. It's a hiring is a is just a a massive challenge and I can't believe that AI's made it easier.
Will (29:08)
But that's why the personal like to your point, John, like I I mean when I've hired, you know, if you have a hundred applications but someone on your team has some point of contact with one of them, you know, maybe they took the time to reach out. It's like and they're like, Yeah, they seem okay or right, like that person goes top of the queue because
Even pre-AI, you can never really filter it. You know, you always want some sort of human validation, right? Because then it's if someone you trust says something, then you're sort of evaluating their them on it too, right? So
Jon (29:38)
Yeah.
Will (29:39)
you always want to find a person at the place.
Jon (29:43)
Yeah, and right now, with the AI thing, there's loads of, I gave a lightning talk at DjangoCon and PyCon this year
Will (29:51)
Mm.
Jon (29:52)
about the problem of job and hiring fraud in job processes. It's global now, it's not just limited to one place, but it's almost...
Joked about going to conferences to meet real people somebody a client of mine came to me said hey John We're sick and tired of like not knowing if the person was speaking to is actually real or not Could you send us profiles of people who you have met and it's like yeah, absolutely like that's that's great and it validates the whole you know, whole trips every time but it's Is that you network of you know first collections it is your
That's the valuable thing. If you're job searching, lean into that. Go and speak to the people that you've worked with before. See who you know is working somewhere who's hiring. Together, you'll find it rather than faceless applications. As soon as you go blind, all the credit that you've got somewhere else is gone. You've got to start from scratch.
Carlton (31:01)
Yeah. So let letting your personal network know that you're job searching is in fact the way forward. It's not you know, you th you think, I don't wanna do that, it's embarrassing, I don't wanna bother people, but
Jon (31:09)
It's... yeah, I mean, it's... yeah,
it's...
You know what, now it's a bonus having a network of people that you've collected that you've met at conferences and things. I'm quite comfortable getting somebody's LinkedIn or connecting to them online if I haven't met them, but I've seen them in the room and just a little note saying, hey, we're at DjangoCon, I wanted to say hello, but you were busy. I don't mind doing that, but some people will hold off. But I think even to a point of going to conferences, going to the
Django on the Med or something like that and sharing the photo that you were there or yeah in Chicago the the silver bean thing that you were
Carlton (31:50)
Yeah, yeah, yeah. I was there.
Jon (31:53)
there and yeah and because it's just leaving a bit of a digital paper trail for people but also
Carlton (31:59)
Yeah.
Jon (31:59)
others will see it and trying to yeah so when you do need to put your hand up
Carlton (32:05)
That's that's how it
Jon (32:06)
and say I'm looking there's people remembering you a little bit
Carlton (32:10)
That digital paper trail thing though, that that hasn't really changed. I mean, I remember twenty years ago the advice was, you know, why why bother answering questions on Stack Overflow? Well, because every single time you did it, it was like the one more sort of thing that you'd done that was positive. Why bother on the on the mailing list? Why bother on the foot on the project? Well, because each time you do it, it's not a big thing, but it just adds to that that that backstory of things that you've done that show that you're, you know.
Jon (32:40)
Yeah, incredible.
Carlton (32:42)
I can't think yeah, that you're credible. Yeah, good. Thank you for the word, yeah, that you're credible, yeah, exactly.
Will (32:45)
You s you still need a I mean you
still need a blog. I I you know, even if AI is gonna steal it, you know, like if I hear about someone, I'm gonna Google them, I'm gonna see LinkedIn, I'm gonna see GitHub. GitHub's pretty meaningless. If they have a blog, I wanna see write about you know, like you you still some most of the advice I think still holds. Like write about what you're learning. Like and doesn't matter where it is, because to me it makes you human. and it also lets me know how you think versus, you know, your resume.
I mean I don't you know after a couple years out of college I don't care where you went personally. I did have a questi a question. so Python web development. I know we all love Django on this show. Are you seeing what are you seeing for Flask and Fast API? Does that come up with clients at all or in the marketplace? Like what are you what are you seeing, John?
Jon (33:37)
I
don't know whether I've flown the Django flag too high, but nobody ever really asked
Will (33:43)
Ha ha ha.
Jon (33:44)
me for any help with flask or fast API roles.
Will (33:49)
Well, I d I don't get the I don't get the sense there are as many, or I get the sense it's like they they
They just want to be able program and then they kinda come in, whereas Django is seen as this a big thing where knowing Django has value,
Jon (34:01)
Yeah.
Will (34:02)
whereas if you can code, it's like, nah, you can sling an endpoint and whatever. I mean, that's my very naive take, but I also don't see anyone asking for Flask or Fast API roles. I see, you know, web programming, yeah, you can learn Python, it's easy kind of stuff.
Jon (34:21)
I read the results of the Django developer survey and those things to see what's coming through. And every year I see fast API is always there gaining ground in the answers about what people are using. But I don't see it on CVs and resumes. I don't see it in job listings. And I don't know whether that's because...
Yeah, people are so happy they're not looking for work because they've got faster API jobs or it's just, it's just, there isn't necessarily that whole ecosystem behind it. But if you get Django's 20, 21 years old, people have been working with it since, I I played the Django developer in like 2008. And that was somebody who was a hobbyist who'd working with Python in their day job and they just.
playing around with Django at home because there wasn't a project to take on professionally. Back then, I didn't see Django on CVs and resumes or anything like that. At that point, you had to go out and look past it. So maybe in the next four or five years, we'll see more fast API jobs specifically, but I don't.
From a careers perspective, I don't see many people switching from Django to FastAPI because they think their work prospects are better.
Carlton (35:53)
Survey is also a bit skewed. I remember the Python survey last year was 50% of respondents have been dip the programming for less than two years. Well, the the community isn't growing at 100% in fifth in two years. So
Jon (36:07)
Yeah.
Carlton (36:08)
who's answered it? And the same with the Django survey, it's like we've got maybe 4,000 respondents, maybe out of a user base of millions. It's like, well, where are the rest of the user base? They're not answering the survey, is the problem.
Jon (36:18)
Yeah.
Will (36:21)
Yeah,
that cuts deep. But yeah, it was about half newcomers and half grizzled veterans on the Django one. And and I know a little bit about the Python one. but it's I'll just say as a quick note, because PyCharm JetBrains sponsors the the PSF Python Software Foundation's survey and the Django survey.
responses are just down in general. People are just not you know, it's not just us, even Stack Overflow, even GitHub, people are just not responding as much. the PSF is not doing their survey next year for a variety of reasons. Hopefully it's a pause. the Django survey we're still doing for now. hopefully it's valuable, but yeah, it's it's completely imperfect measure. It's whoever finds out about it.
And yet it's as good as we have. one thing we can mention, it's it's public is that the Django project site now has very basic anonymous tracking with plausible, I believe. So some sense of who's actually visiting, but you know, who's to say what's a human and what's a bot these days, right? Like that's part of the problem. I mean, I went through I was sort of and you know, bots, bots are nothing new, but the rate has just increased so much. I mean, so I've I've I run this small learnjango.com site, right, where people can
and
tutorials and courses. And I was sort of smugly like, I've got a good number of users and it's really accelerating, you know. And then I did some analytics on it, you know, and yeah, it's like all bot not all bots. It's like the s the number of people is pretty consistent in terms of growing and it's the rate of bots is just through the roof. And as soon as I've you know slapped on some ways that bots can't sign up as easily, like it dropped like crazy, you know. So
Yeah, it's to your point, John, it's you still need humans and it's harder than ever to know who's actually human out there.
Jon (38:10)
Yeah, but I think with the, you said like the drop off on the surveys and things, think notably like the last couple of years, the numbers of people at the conferences is maybe lower. I don't know if that's, what's driving that? Yeah.
Will (38:20)
But it's not it's not Python Django is what I'm saying. It's not Python Django. It's
I hear that across all all the places. I mean, I think it's I I'm curious what you think. What I've heard is, you know, so there's travel is harder, there's political stuff, budgets are tighter, more money's going to AI, people are working from home. I ha I've heard it's just kind of everything. And this is from conference organizers too. I don't know that there's one single thing causing it other than it's all hurting.
Jon (38:51)
think organising conferences is hard work, so it must be even harder now when you don't know how many people are coming and if a lot of the sign-ups are either very early or very late, it must be very
Carlton (39:04)
Yeah.
Jon (39:05)
difficult.
Will (39:07)
Yeah.
And yet the value is there, right? I mean, I I think I met you at a conference, John. I certainly met Carlton at a conference.
Jon (39:13)
Yeah.
Will (39:14)
Like this podcast came about be you know, so and there's opportunity grants. So, you know, it bears repeating, like, challenging as it is to go, it is so valuable to go. And, you know, I think especially if you're a junior or you're new to the community, you get so much out of it because people are friendly and helpful and
You know, the people I've I met at DjangoCon US in Chicago last year, I was re you know, referring you to John. You know, you y most people are very friendly in the community. So if you can find a way to go, it's probably the best bang for your buck if you're trying to, you know, get out there, I would say.
Jon (39:51)
It'll be interesting to see next year with DjangoCon going to California.
Will (39:59)
Yeah, Riverside.
Jon (40:02)
If people haven't been coming to the East Coast for the last couple of years because of the travel, if that makes a big difference. thought the ones in San Diego were very well attended even during those Covid years.
Will (40:19)
Yeah, we'll see. so we have a books and projects section. Carlton, were there questions you wanted to ask
before?
Carlton (40:28)
No, no, no, it's all
good. I mean I the one thing I might just mention was Django dot social, which you kicked off John and that's a yet another good way that l people can organise a local thing. Get together, chat about Django, meet peop local Django developers and and whatnot.
Jon (40:44)
Yeah,
yeah that was a, I suppose a born out of a tender, I went to, so it must have been like, I'm trying think what the dates were. April 2022, I went to Django Day Copenhagen, which was like the first in-person event for.
years and I went there and it was only 40 people, 50 people, but it was great. And it was like I'm out and about, I'm with the people that I enjoy spending time with in person and I came back and I said, right, I want to try getting something going up in London again. it was like, yeah, just I don't want the hassle of having to find venues and speakers. So.
saying right let's just meet and do something became very easy.
And then classically I set up on meetup.com and ran it through there for maybe two years or three years. Set up extra additional events in and around the conferences and traveling and then people would take it away and go, right, I'm gonna go set one up wherever they are in the world. And we ended up with a network of maybe about 10, 12 events dotted all over the world. And then I started getting bills through from Meetup for $3,000 a year. And it was like,
Carlton (42:00)
Yeah, it's a bit much.
Jon (42:04)
this is a nice thing to do, but it's it's not $3,000 a year nice on top of everything else, and
Carlton (42:11)
Yeah, yeah.
Jon (42:12)
most of these events happen once and then don't happen again. So I took it all off Meetup, and it sort of not died, but it doesn't really happen as much anymore regrettably. So I've sort of...
I've mostly been trying to do it as like a night before the conference social event wherever I've
Carlton (42:35)
Yeah.
Jon (42:37)
been. So we did it in Athens and there's quite a few.
Carlton (42:40)
And in Dublin.
Jon (42:42)
And Dublin, yeah Dublin was good and I intend to do one like an out-praying one in Austria or maybe on
Carlton (42:48)
Alright, yeah, yeah, yeah.
Jon (42:49)
the, yeah somewhere near the ski lift. so yeah, but it's a great way to just socialize at a conference now.
But if somebody wants to run one, feel free to use the name and crack on with it. But running meetups is difficult.
Carlton (43:04)
Yeah, yeah, brilliant.
Well yeah, we need to vibe code some little Django app to host it rather than using meet
Jon (43:10)
Yeah.
Carlton (43:10)
meetup.
Will (43:11)
Yeah, I mean Meet Meetup decided
I mean, same thing here in Boston, like a c a number of years ago. They decided let's just jack the price insanely high. We're gonna lose ninety five percent of our users, but we'll just milk the five percent. I think there was new ownership. So yeah, it's keep yeah, I mean, in theory someone can vibe code something up and host it and
Jon (43:30)
Andrew Miller in Cambridge still runs one
Will (43:33)
Mm, yeah.
Jon (43:34)
every month pretty much. So he and I sort of co-hosted the one in Athens.
he's found an app that he was using to manage it locally. So we just ran it as a Cambridge one and I'd stuck someone out in the Slack channel a couple of days before. it's easy. That's the thing,
Carlton (43:57)
K
Jon (43:58)
the whole idea of Django Social was it should just be so low admin that you just pick somewhere to go and especially when a conference is happening in a big city and you can be...
you can go out and not bump into anybody. It's useful to say this is happening here.
Carlton (44:14)
Yeah. Yeah, no, it's really good that you organise something and you get yeah,
we're we're gonna be here, you don't have to come, but and then you obviously get hundred people because you know I think the the bar in
Jon (44:20)
Yeah.
Carlton (44:25)
Athens they were a bit like, Whoa, what is going on? They did they weren't expecting us.
Jon (44:28)
We'd already been to one bar that we'd originally
said we were going to run it here in X bar and we got there like, no, we can't, you can't do that. You can have a table each, you can't have 100 people or 50 people on
Carlton (44:42)
Right.
Jon (44:42)
one tab and they were like, it just didn't work. So we just panicked and went to this other place and they're like, yeah, whatever, we'll just serve
Carlton (44:49)
Yeah.
Jon (44:50)
you at the bar.
Carlton (44:51)
That's good. I think there's a lesson in in in that. It's like, you know, Django being such a small community, to keep things simple and run them small and don't y there's always you know, this I know you just set it up on on meetup and then it grew, but d quite often we have a tendency to try and grow the thing to the next level. It's like, No, keep it small, 'cause then it you can keep doing it.
Jon (45:08)
Yeah.
Will (45:10)
Well, speaking of which Django and the med, you're all you mentioned before, but you're already at pretty much capacity for size, right? You think going forward?
Carlton (45:16)
Well.
We yeah, so we
had twenty five people this time and I think the twenty to twenty five range is about the right size for it with the vibe that it has. And I think if we were to grow it beyond that, we'd have to change the format somehow. And it's like okay, so I think Jang on the Med needs to be you know, or a small sprint event 'cause I think the the the hope is that people will pick up and run with them elsewhere. But twenty, twenty five people that's a good number because you can fit in a room and you can just about get a table at a restaurant if you have to and y you can get to
bar without the you know people panicking. Whereas if it's fifty or a hundred, well okay, you need a much bigger venue and much bigger organization and you can't do the the group activities in the same way and and whatnot. So yeah, I think there's a natural size to a sprint event like Chang on the Med.
Will (46:03)
Yeah. Well and there at DjangoCon Yeah,
Jon (46:04)
have to run more of them.
Will (46:06)
yeah. Well DjangoCon
Carlton (46:06)
Yeah, exactly. More.
Will (46:08)
US there was a lot of talk around different venues and places, so I can't speak for other people, but I think there'll be more it doesn't have to be on the med. Always. books. Should we John, we didn't mention this to you, so you're off the hook, but Carlton and I like to subject people
to books that we're if you have one though, please.
Jon (46:27)
It's alright,
I listen enough and I know that you always ask this so I was panicking because I pretty
Will (46:34)
Do you want
Jon (46:34)
much only read Jack Reacher books. You've
Carlton (46:38)
Okay, you recommend one of them?
Will (46:39)
Okay. Okay, yeah, what's w
Jon (46:42)
read one, you've read them all, it's fine. Any one of them, you know exactly what goes on. But there is a book that I really...
took a lot from and it's called the GoGiver and it's about the, it's a business theory and it's...
about the they they they do a lot of theory in it and then they have like a story in the middle that helps you to understand it in like real context and it's mostly about going in not going into situations expecting a result and just going into it fully open-minded and it's almost like where what has driven the way that i think about going to conferences and
involvement in the community. Nothing has got like a what's in it for me sort of mindset. It's all if I go here it's the right thing to do and think yeah good will come from it, good will happen. So yeah it's called I think it's called Go Give Us Selmore but yeah and often as a if I recommend it to somebody I will buy it for them as well so I will send you both a copy. Just that's the nature of the book.
Carlton (47:56)
wonderful one.
Thank you very much.
so I've been reading, rereading Gilbert Rile's can you see that? Hang on. it's The Concept of Mind anyway by Gilbert Rile, which this is the Penguin Classic edition. it's a sort of classic of 20th century British analytic philosophy. But in this these days and ages, this day and age where we have tech billionaires writing manifestos about, you know, cognizant AI rights and how reinforcement learning is tantamount to torture and these kind of things.
It seemed important to me to refresh myself on the arguments around, you know, how we conceptualise mind and intelligence and rationality and other such concepts. So I'm I'm rereading that. I'm not sure I'm gonna have another book for the next episode will, because you know, ph philosophy books are quite slow going, so I'm I might not have plowed my my way through it, might still be on it.
Will (48:52)
Yeah, well that's okay.
Carlton (48:53)
But but that's but I'm so I'm reading that. So Gilbert Ryder's concept of mind, which is as I say, a classic from twentieth century.
Will (49:02)
New to me. I'm reading hold it up. The Infinity Machine, Sebastian Malaby about the Demis Has Hasabas, the Deep Mind founder. And I've read other books, you know, I I like reading nonfiction books around AI. I mean Empire of AI, I think, was the best one I read last year. but I find this interesting because AI is much bigger than LLMs, and DeepMind has done a lot more than just LLMs, right? They're Alpha Go, they're ch you know, and
chess engines and reinforcement learning and so I find it interesting looking at things beyond LLMs, because AI is much bigger than that. So I'll let you know how it goes. I've read other stuff by the author, so hopefully it's interesting.
Carlton (49:47)
You you have got an amazing stamina for these nonfiction AI books, Will, I must say. I must congratulate you.
Will (49:53)
Well, I find people interesting and, you know, I'm I am a history major, right? So I like to know the context and the motions, you know. You know and I would I would repay you the same I think it's a compliment that your stamina for philosophy books is amazing. So I I f I I've
Carlton (50:08)
Yeah, no, I I love him. I love him.
Will (50:10)
I feel like I can like digest I'm like two, three sentences, I'm like, Yep, that's two, three hundred pages, but I know it's about the process of
internalizing the the debate. projects, I'll go. I'm gonna mention Justin Mayer's Django Benchmark, which was started at Django on the Med, is now public. Carlton, you were there with Justin when he w announced
it
Carlton (50:33)
Yeah.
Will (50:33)
or made it public, right? Wha wha you can you add anything?
Carlton (50:36)
Yeah, so him
and Jeanne Andalay and then Lillian also was helping on it at the sprints. They so what it is is a harness for testing different Django different kinds of Django application under different deployment scenarios. So that at the moment it's very much path in the snow, you know, it's beginning. But i we the the the the the harness is there, the ability to add new s scenarios is there. And for instance, you know, you your blog application. Well, you know, how di how does it go running under Gunnarcorn with
you know, a a pre forked worker versus a threaded worker versus well what about if you run it as as a as an ASGI application and how does that perform at this at the scales that you're interested in. So we also bit you know also load test the application and you like, well, you know, it it might fall over at 500 requests per second, but yeah, I'm not getting five hundred requests per second to my blog. So do you know what my my Garnicon worker with its two threaded workers is perfectly fine. the the the idea is to be able to talk a little bit more
more sensibly about how you should deploy your Django application than just like, look, it does so many requests per second, which is a little bit, you know, vacuous basically. It's like when you know, I want to deploy a Wagtail app, I'm expecting this kind of traffic. Okay, here's a here's a suggested deployment that comes out from running it in all these different ways.
Will (51:58)
Yeah, it's exciting. So hopefully it grows.
Carlton (52:00)
Yeah. I d I think it's been it's been something that's been on the the sort of, you know, idea pot for a a long time and to see it coming together I th I'm very excited by Django Benchmark. I think it's gonna be a very good un community resource, you know, as it as it enriches. So, you know, I think if people have got people want to sort of add their application to it, that would be a useful thing so we can build up the the the s the different scenarios or the different, you know different kinds of application that are being tested.
That's good.
I want to just mention an application by Fahan Aliraza, who's the you know, is the Google Summer Co candidate who added Django template partials into core and who the author of the Django Bolt project. Well he's also which is the super fast Rust, you know, application server for Django. he's come up with a great package now called Pongs, which a prongs, sorry, which runs just the tests that you edited. So you know, your full test suite might take two, three minutes to run or whatever.
And you're only working on two or three files, well it it'll watch the files that you've editing and it will just run the tests that are relevant to those. And that's that's just so useful. I've Yeah.
Will (53:10)
It's so so smart. So smart. Yeah.
Carlton (53:12)
Yeah. No, I I'll I of I had a half idea to do similar with coverage data, but he's gone and actually built it and of course it's much better than anything I would have come up with. So good do go and check that one out. That's Farhan's prongs project, which will it link in the show notes.
Will (53:28)
John, don't feel excluded, but also don't feel obligated to add a a project to this.
Jon (53:33)
The
only project I've got eyes on at the minute is trying to the podcast going.
Carlton (53:41)
Okay, so people are code the project podcaster.
Jon (53:43)
People are kind,
yeah.
Will (53:45)
Yes. And then is that just on YouTube or is it are there also audio links for the
Jon (53:49)
I managed to
get it on Spotify yesterday and I'm just trying to figure out how to put it onto Apple Podcasts as well. So yeah, it's intention to be
Will (53:57)
Okay. Yeah, I was looking. I th f
Jon (53:59)
a cross-platform, yeah.
Will (54:01)
Okay. Great.
what haven't we asked you, John? Was there anything you wanted to get off your chest? I I would say my my wife told me, she's like, I got a great joke for you and I was like, Okay. And she's like, that you know, the deepest, darkest secret in the world you put at the end of a podcast. And she thought that was hilarious. And I was like, I heard that when we started doing this podcast, but it's still funny. But, you know, so you can get anything off your chest now. If anyone's still listening, you know, it's any hot takes.
Jon (54:31)
I tend to share those anyway,
don't have to wait till the hidden minute at the back.
Carlton (54:34)
Ha ha ha.
Will (54:37)
Okay, but people should definitely Foxley Talent, LinkedIn, they should listen to the podcasts and if they're going to any of these events, you are a ubiquitous you know, member of the community as well as supporter, which everyone should do and people shouldn't be shy.
Jon (54:54)
Yeah, no, absolutely. Just come and say hello. That's the whole point on that. Just to meet people and integrate and be around. yeah, no, Any question answered, ask me anything like those, yeah. What's it called? Like coffee hours or those sort of things. Yeah, just grab us. I'm easy, easy going.
Will (55:15)
All right. Well, John, thank you for taking the time. Nice to formally do this.
Jon (55:20)
Yeah. Yeah.
Will (55:22)
And we'll have links to everything and we'll see everyone next time. We're back on the usual schedule for anyone worried. So we're gonna go every two weeks, I think, through the end of the year, Carlton, right? We got a whole bunch of guests lined up. So we're back.
Carlton (55:32)
Yep, that's the plan.
Will (55:35)
All right, we'll see everyone next time, DjangoChat dot com. Bye bye.
Carlton (55:39)
Bye bye.