i do email marketing for a living - not frontend.
so i built this instead: tap any part of a 2026 f1 car and it tells you how it works, what it costs, and why races are won by it.
13 parts. built on nights because i love the sport. €12–20M of machine that weighs 768kg.
what should i break down next?
@SaraDiscovers don't listen to the noise, although gpt 5.6 limits are very friendly compared to fable 5, good luck running out of them on max plan before reset hits.
We've just added several new features to Claude Managed Agents.
You can now configure effort levels per agent, seed sessions with events, add up to 500 skills per session, use webhooks for environments + memory stores, and stream events for sub-agents.
i do email marketing for a living - not frontend.
so i built this instead: tap any part of a 2026 f1 car and it tells you how it works, what it costs, and why races are won by it.
13 parts. built on nights because i love the sport. €12–20M of machine that weighs 768kg.
what should i break down next?
confession. i had code running for a week before i could clearly say what the value prop even was
that’s backwards and i knew it while doing it. building got so cheap it’s now easier to ship than to think
the bottleneck moved. it’s not “can you build it.” it’s “should this exist”
@MaxHirsch13 I don't like it, vercel + supabase it's too common, personally I prefer any vps, hostinger, hetzner, namecheap etc. There's also https://t.co/q75T0t166s
"Gemini 3.6 Flash is the exact same model as 3.5 Flash"
Same Intelligence Index - 50 and 50. Sure.
Hallucination rate:
3.5 Flash - 61%
3.6 Flash - 54%
Same intelligence, meaningfully less confident nonsense.
Everyone benchmarked the wrong axis and called it a flop.