In conversation with Shawn K
From being replaced by AI to embracing it
It was a blast chatting with
on Friday (apart from the inevitable WiFi issues on my end).His remarkable journey from being replaced by AI to going viral on Substack, and now running an AI-first team as Director of Engineering, is genuinely inspiring.
Shawn has some great insights into what the engineering team of the future will look like (and what it looks like today).
We spoke for almost an hour, but it felt like we were just getting started!
Check out the full conversation below.
As always, here is a summary if you don’t have time to watch or listen.
And make sure to sign up to Shawn’s Substack if you’re interested in getting practical tips and future insights into building with AI.
AI Summary
This conversation spans a critical inflection point in software development. Shawn’s story—unemployment, a viral post about AI’s impact, unexpected support, and finally a company willing to take a chance—frames the larger question both hosts are circling: what actually remains valuable when AI handles the coding? Rather than debating job elimination, they explore what distinguishes exceptional engineers. The answer isn’t about SQL or specific languages or even architectural knowledge. It’s about taste, about understanding customers, about making deliberate trade-off decisions between shipping today and building flexibly for tomorrow. Their discussion reveals that as the technical barrier to software creation collapses, the skills that matter—empathy, judgment, problem-solving—have never been more central. This isn’t a conversation about adaptation. It’s about recognising what was always at the heart of good engineering.
Chapters
(00:00:01) - Live stream launch and technical difficulties
(00:01:04) - Alex’s fractional CTO practice working with startups
(00:02:20) - Shawn’s 15-month unemployment and AI’s unexpected role in it
(00:03:18) - When a personal story goes viral: the consequences and support
(00:04:42) - The AI paradox: everyone thinks it’s amazing except at their job
(00:06:00) - MVP is dead: why “Minimum Useful Product” actually matters
(00:15:00) - Rethinking how junior engineers should learn in 2025
(00:30:00) - Don’t bother learning SQL: reframing technical skill priorities
(00:46:00) - The eternal trade-off space: shipping today vs. flexibility tomorrow
(00:48:00) - Imagining engineering when software itself becomes solved
(00:49:13) - Taste and judgment: what experience can’t be automated
(00:51:29) - Selecting for character, not just technical ability
Key Moments
The AI Paradox (00:04:42) — “Everybody wants to believe that AI can do everything except their specific unique role.”
Getting the Chance (00:02:46) — “They took kind of a risk, because I was talking so big about AI solving all our problems, and they took me up on it.”
What Remains (00:49:17) — “What’s going to remain is your taste and what you’re deciding to go after. More experienced people are still probably going to be better at that in the future, even when software is solved.”
Why Junior Engineers Still Matter (00:51:29) — “Those are the kind of juniors I select for. They have the chops of working with software but more importantly, talking with clients, figuring out, solving real problems, delivering genuine tangible products.”
The Truth at the Core (00:52:21) — “Your job as a developer... it’s less and less has to do with code and technology. You are a problem solver. That’s what should be internalised.”
Sharing is caring!
If you enjoyed this conversation, please consider sharing it with two friends who might find it useful.