Hi, I'm Larry.
I'm an Australian software engineer based in London. I work at CO2 AI, building tools that help large companies measure and cut their carbon emissions.
Most of what I think about these days is how our work is changing now that we build software with AI. Over the last year I've helped my team go from a couple of us experimenting with coding agents to most of the team using them every day. I've been running training, building internal tooling, and dealing with the new challenges that arise when everyone is using agents.
Things are changing so quickly that we need to be adaptable and open to change. Anyone who says they have the perfect workflow is underestimating how things will continue to change. So I try to keep up to date with the latest trends, find the patterns that will last, and drive change in my team.
This site is where I share what I'm reading, what I'm building and what I'm thinking. Here's the latest.
📚 Reading
“Fiona Fung (manager of the Claude Code and Cowork teams) on Lenny’s Podcast”
Really thought-provoking interview, and great to hear how the Claude Code team is dealing with challenges that come with AI engineering. Anthropic and OpenAI have the advantage of unlimited tokens, but are worth paying attention to as they point towards the future of how software teams build.📚 Reading
“Stop Taking Tickets, Start Applying Jobs Theory”
A useful lens for engineers to apply when scoping a piece of work - based on the work of Clayton Christensen. Going to try this out!📚 Reading
“Some things I’ve learned recently running coding agents on large-scale projects”
Basically be more ambitious, queue up lots of work, and enable agents to close the loop. Need to try following this advice!📚 Reading
“Most software engineers are facing an identity crisis bordering on depression.”
This is an exaggerated example but touches on something true. This is a challenge that needs to be solved by the engineering culture of a team.📚 Reading
“Loop Engineering”
Great overview on setting up automatic AI workflows. A good nudge for us to stop babysitting agents so that we can spend more time where we actually add value.📚 Reading
“Learn anything with the /teach skill”
Matt Pocock is the king of agent skills right now. This video itself is a great piece of teaching about how to make skills, and the skill itself looks great! I’m going to try to learn how LLM inference works with this.📚 Reading
“Why AI hasn't replaced software engineers, and won't”
A good counter to the hype that AI means mass layoffs and the end of software engineering.📚 Reading
“AI enthusiasts and skeptics are running different races”
Charity Majors on treating AI adoption as a leadership and engineering challenge, not an argument.