Archive · July–August 2026
The five AI talks worth your time, week by week.
— and the exact minute to jump to.
Hours of AI talks drop every week. Through the summer of 2026 I monitored the channels that matter, cut the hype, and picked the five worth watching — with what you'll learn and where to jump. No new issues ship; the archive stays.
FounderTLDR is no longer publishing.
The last issue went out in August 2026. The five issues and every talk write-up stay here, free, for as long as the archive is useful.
How Unify cut its AI agent costs 95% in two weeks
Connor Hegy · LangChain · 1h 8m
A CTO walks through the actual engineering behind a 90-95% agent cost reduction in two weeks: collapsing many sub-agents into one code-writing main agent, working around OpenAI's 15 req/s prompt-cache limit by hashing users across 16 buckets, and building a zero-cost TypeScript sandbox. It's rare to get this level of production detail on agent economics, including why they optimize tool-call efficiency over per-token price.
Read the full issueLatest issue · Nº 005 · Aug 11 – Aug 18, 2026
3h 37m of talks → one 4-minute read
How Unify cut agent costs 95%, and why RL-trained agents break in the wild
- 01How Unify cut its AI agent costs 95% in two weeksConnor Hegy · LangChain1h 8mwatch
- 02Chelsea Finn: This is the State of the Art in RoboticsChelsea Finn · Y Combinator58 minwatch
- 03From RL to IRL — Gaurav Mishra, Amazon AGI LabGaurav Mishra · AI Engineer18 minwatch
- 04Exo: Harnesses should see their own code and logs — Alex KrentselAlex Krentsel · Latent Space47 minwatch
- 05How to Cut Your LLM Classification Costs by 90%Hamel Husain26 minskim
The filter
Substance over hype. Every single week.
Most AI newsletters amplify whatever is loudest. This one does the opposite: primary sources, technical depth, real insight — or it doesn't make the cut. If a talk is a product pitch in disguise, it's out. If a week is thin, I'll say so instead of padding it.
— Héctor, builder & curator
CURATING FROM Anthropic · OpenAI · Google DeepMind · AI Engineer · Latent Space · Stanford Online + 35 more channels
How it works
01
Monitor
41 hand-picked channels — research labs, conferences, builder podcasts. Every new talk lands in the pipeline within hours.
02
Score
Each talk is scored for substance. Clickbait framing, news rehash and product marketing get cut before I ever see them.
03
Distill
I pick the five that matter and write why — what you'll learn, and the exact minute to jump to. Five picks, one four-minute read.
Five talks. Four minutes.
Five issues.
FounderTLDR stopped publishing in September 2026. The five issues and their write-ups stay here.
FounderTLDR is no longer publishing.
The last issue went out in August 2026. The five issues and every talk write-up stay here, free, for as long as the archive is useful.