Building | a16z speedrun scout | Co-founder and former CTO of Dex
London, England
Joined August 2011
- Tweets2.3K
- Following985
- Followers405
- Likes8.4K
Pinned Tweet
Very grateful to have been named in the Forbes 30 Under 30 Europe AI list!
It was a surprise seeing this come in right as I was stepping back from Dex, but I'll admit I was very happy to see my achievements marked - building a great team and the best voice AI experience in recruitment, and raising $8.4m in funding.
Really though, it was a great reminder of just how early in my career I am. So stay tuned, the best is yet to come.
And yes, I promise this is the only time I'll post about it!
absolutely nobody could have predicted this
Anthropic is now recommending that companies REMOVE unnecessary subagents from their AI systems.
This is a pretty significant shift in agent architecture, with their team finding that a single agent equipped with skills can frequently outperform complex multi-agent systems, and one customer seeing roughly 2x faster execution.
The emerging rules of thumb:
→ Skills: Teach the agent HOW to do something, including procedures, expertise, and instructions.
→ Tools / MCP: Give the agent the ability to DO things through APIs, actions, and live data.
→ Subagents: Use for parallel work, isolated context, or tasks a single agent struggles with.
The interesting part is progressive disclosure. Instead of spawning specialized agents or stuffing everything into a giant prompt, one agent discovers and loads skills and tools as needed.
Their advice is to fold existing subagents into the main loop, turn their instructions into skills, expose their actions as tools, and run evals to see if performance holds up.
As models have gotten smarter, improving agent architecture is becoming less about what you can add and more about what you can delete.
claude.com/resources/article…
I'm analysing someone's app to help them with a migration. I didn't ask it to look for security vulnerabilities, but Claude found some, then flagged the session for cyber risk and shut it down 😂🤡
based af. wish more artists took this stance. lecturing your audience between songs achieves nothing and alienates half the crowd
Without spoiling it, I have a slightly different take on Digger. As far as I can tell most of the people who like it seem to be having their cake and eating it. They like both the message and the meta-commentary around it. Haters wanted one or the other.
I also really liked it, but came away feeling that the message could have been anything. I felt it was asking what your Level 4 was. Anyone else?
While I am loathe to do LinkedIn posting on X, I should probably share here that I'll be speaking at Venture Intelligence Day in Prague next week 🇨🇿
I first spoke at VID two years ago where I gave a deep dive on Atomico's data-driven VC stack. Nowadays most funds are comfortable with data, and many have successfully experimented with AI tools. Fully integrating those tools with their data and preventing new silos emerging remains a challenge.
In my talk I'll be discussing:
- What it really means to be an AI-native VC
- Practical examples you can deploy at your fund
- How to turn AI experiments into a firm-wide advantage
Let me know if I'll be seeing you there!
Original post link: lnkd.in/p/efuN6UDh
SNL UK’s Weekend Update is simply electric. Consistently hilarious week after week and getting noticeably braver too, love to see it
Finally saw the Odyssey on Wednesday. Very mixed feelings. Thought I'd go watch it again, but think I've finally decided it wasn't for me and that's ok. (Sorry)
1) i don’t think it was great that the ‘launch’ of Astra left us uncertain when/if it would roll out
2) rolling it out to everyone this quickly is awesome, well done to the OpenAI team
