For the next 2 weeks @casualzach_ and I will be live streaming the build of a self-driving product in the open entirely automated with agents using @ona_hq.
Follow along at: @swfactory_dev
All details at: https://t.co/riP0XBoDO6
(many more updates to come!)
@leerob > The SF hivemind has a tendency to overfit what works at startups for every company.
Yup. After working at startups specialized dev roles seemed very pointless to me. Then I joined some big corporates and I was the weird one.
@Biohazd5 I don't think we did, and it would be quite custom to DAZN as it was converting many of our internal plugins etc, and using specific patterns. That said... feels like this would have been 1000% easier if we had LLMs :D
Today, I walked past the Old Street roundabout on my way to a SXSW London side event, and saw @synthesiaIO on a billboard with some of the best companies Europe has ever produced. I'll admit it gave me a much-needed energy boost ahead of speaking on yet another panel about AI's problems.
The billboard is part of #BuiltinEurope, a campaign @balderton has launched to tell the truth about what is happening right now in European tech.
The story we've heard for years has been one of catching up. Europe could lead, the headlines said, if only it had fixed this or that. That's because building in Europe was treated as a thing you had to overcome.
Now, for the first time in a long while, the opposite is true. Met with indifference at best and opposition at worst at home, European founders are forced to build companies that work everywhere from the first day, and they're doing it with style (and substance).
At Synthesia we feel this in our bones. We were built in London and we have been global from the start, because being European made that the only sensible way to think.
So if you have ever thought about building a company in Europe, or joining one, consider this your invitation. https://t.co/cvFiwU9dG3 is now live, with a jobs platform pulling together open roles from a thousand European startups, and a directory of the continent's best incubators for anyone ready to start something of their own.
Europe is building. Come build with us.
Get to know River, our internal agent that among other things drastically reduces the cost of context gathering across the company. No more digging through docs, Slack, and requiring people to spread context. A new kind of information architecture.
Talked about this with @tobi on our Context episode:
my "plans" largely look like pseudo code composed of mostly types/interfaces, how they compose, and their boundaries
ive recently started including call stacks - been very helpful for both me and agents when implementing
Most teams think they're using background agents. They're using coding assistants.
The difference: close your laptop and see if the work stops.
Background agents run in the cloud, trigger from events, and open PRs while you sleep. Stripe ships 1,000+ agent PRs/week this way.
We at @ona_hq wrote the guide: https://t.co/vhjWGIC7il
@ddunderfelt@jamonholmgren It's a good point, but some low-touch coaching would go a long way. Some internal stakeholders will be overbearing, but most should be amenable. That even said, the cost of 'overdelivering' would be fairly low. It's less problematic to gold plate internal apps.
My current gut feel is that juniors will start at small companies as the first or second developer hire, for developing in-house software.
They’ll build low stakes apps for those companies, like estimating, automations, websites, task management.
It’ll teach them the strategic layer (since they own the whole stack) in a lower pressure environment. They’ll need to think about all of it, but the projects are small and the user bases tiny.
This is as opposed to being a junior on a larger tech team, where you’re given small tactical slices to work on under close supervision. That route is drying up, I’m sad to say.
These juniors will be affordable to a small business and AI will give them instant productivity that the business can leverage for operational efficiencies.
My son @cedricholmgren is currently doing exactly this route; working at a local fence and deck company as their in-house programmer. He owns probably a dozen apps in 5 different languages and frameworks. It’s amazing. He’s been doing it for a couple years and at some point someone’s going to discover that he’s a lot smarter and more disciplined than his old man and snap him up.
Graduating on then to user-facing applications makes sense, this also allows you to learn the ropes in product type skills like gathering feedback, iterating etc in a low-stakes way with more friendly users.
It's actually a good point - and something I've thought about a lot recently. The idea of juniors working on internal applications and internal process seems quite viable as an approach. As there is now much larger demand for bespoke and internal software, than ever.
My current gut feel is that juniors will start at small companies as the first or second developer hire, for developing in-house software.
They’ll build low stakes apps for those companies, like estimating, automations, websites, task management.
It’ll teach them the strategic layer (since they own the whole stack) in a lower pressure environment. They’ll need to think about all of it, but the projects are small and the user bases tiny.
This is as opposed to being a junior on a larger tech team, where you’re given small tactical slices to work on under close supervision. That route is drying up, I’m sad to say.
These juniors will be affordable to a small business and AI will give them instant productivity that the business can leverage for operational efficiencies.
My son @cedricholmgren is currently doing exactly this route; working at a local fence and deck company as their in-house programmer. He owns probably a dozen apps in 5 different languages and frameworks. It’s amazing. He’s been doing it for a couple years and at some point someone’s going to discover that he’s a lot smarter and more disciplined than his old man and snap him up.
One of the underrated reasons Linear is so popular with so many people is they have an internal target that nothing in their interface should take more than 300ms to render. They keep fixing regressions whenever it happens.
It’s very hard to retrofit this culture: look at JIRA…