
Physics @Cambridge_Uni, ex-VC, building @AISEA_builders. Codex Ambassador.
Malaysia
Joined May 2024
- Tweets424
- Following282
- Followers412
- Likes820
last week I got to give a workshop at the cradle LIVE summit on how to use codex to run a one-person startup
tldr:
- own high leverage tasks where your personal touch gives outsized returns, and handoff any that aren't to an agent)
- manage context properly so your agents know enough, but not so much that it overloads them. set up a structure for your records, breadcrumb and make notes throughout for your agents, and use subagents to keep your main context clean
- set up evals, so that your agents can work reliably 24/7
- make it seamless to interact with your agent. I connected Dots to a system of records on the cloud so I have a secretary on the go with full context of your work
next session in the comments
last week I moderated a fireside chat (ft. @stripe, @kelas_sekejap) on how to think about charging for your app
the tldr is to gather data as fast as possible, and focus on building the funnel and pipeline of users to test out pricing and payment models
identify what your core value is in a single sentence, surface value through your features, and immediately set a paywall once that value is shown, so you can properly see if you're building something people want
after the whole waze/tuju debacle the malaysian govt has been talking about open tender for navigation apps when a mini hack seems like it would do the job better
so as almost a joke I set up a luma event for this
within the first 20 mins before I even started marketing we had 15 signups
so I guess we're doing this now, if you wanna build a better navigation app, join us in the link below
at our last paper club we talked about how to build an agent that improves itself
key takeaway is that you can't really yet
humans build better things because we have access to ground truths, where reality tells us if we're wrong, but AI relies on checks and evals to know it’s going in the right direction
when the check is ground truth, like a game result, a compiler, or a passing test, AI can go well past where it started, but when the evals are written by AI they can only reward what AI already knows how to recognise.
so “self-evolving” usually means evolving towards a targeted preference, not getting better
opportunity is in models that can check against ground truth, or explore beyond what they already know
1
8
312
Paper Club tonight talking about self-evolving agent harnesses
Key takeaways:
- "self-improving" harnesses is just iteratively getting towards a target criteria, and using some constrained gradient descent for iterative improvement seems to beat out spray and pray .mds
- human in the loop is still needed where goals are fuzzy (eg marketing, content) and sometimes the balance is how cheap you can keep RLHF
1
124
malaysian builders need rooms to lock in. so we're opening up ours.
being part of the malaysian.ai residency took us from concept -> pivot -> pivot again -> six-figure revenue in 4 months
we're opening up our first official cohort now. it's 8 weeks, full-time, in person in KL with some of the coolest builders in town
applications close oct 1. if you're obsessed with a problem, apply: platform.malaysian.ai
















