Jay Lee

[michelle @michellechen]

one thing that made this click for me — with decision models, you have to define categories and it returns probabilities that the input matches those categories.

but so much of the data we interact with is completely new to us. how can you define categories for something you've never read?

with clairic, you can extract values directly from the inputs, without any premeditated categories. the pipeline becomes clairic to extract values → clef to make decisions with those values → LLMs to take actions on those decisions. all of these models run on Workers AI.

[Erik MacKinnon @erikmackinnon]

hey in case you didn’t know

@get_bb_app is frickin’ awesome

grateful for the team building this.

thank you!

https://t.co/U50jxvTl5I

[Jay Lee @joiedevivre01]

@petergyang @bot @spacexai @kunchenguid @poteto As a note, all prompts displayed on the mindmap were written by Lauren and are properly cited in the notes for each node.

[Jay Lee @joiedevivre01]

@petergyang @bot @spacexai Building my SaaS app with Firstmate @bot (@kunchenguid), dr eggbot (@poteto), and Cursor Cloud Agents! 🚀 Mindmap is still a WIP, but excited for what's ahead.

[Cary Scardina @scardinafineart]

I designed mine to protect the elderly from email-based scams. It scans for phishing and social engineering indicators, checks email headers, links, and attachment hashes against trusted opensource tools. It is also the gateway to the web for other bots, pre-scanning domains and downloads before my other bots access them. https://t.co/yNZvAzipRx

[Jay Lee @joiedevivre01]

@bot I want that https://t.co/8D6ioO6fWd email address. follow up until i get it.
### Instructions
1. Check the status of the domain/email provisioning step-by-step. 2. If waiting on external steps (DNS propagation, admin approval, account setup), log the current status and timestamp. 3. Prompt or retry every 10 minutes. 4. Stop tracking and notify me immediately once the email address is fully provisioned and ready for use.

[Jay Lee @joiedevivre01]

@wuweiweiwu Mo

[Daniel Hauschildt @DrHouseShield]

From today on you can install @getcodesign directly via claude code 🥳

[Alex @alexgreensh]

i've tested ASD-STE100 in my own evals a few months back. It was great for some sessions and very bad for others - especially knowledge work.

Kept testing and made output styles that worked best. Some use several rules inspired by ASD, and they give me great results without impacting work quality.
Open sourced it here, take a look would be happy to hear your feedback!

I might rerun the benchmarks with Opus 5.5 and Sonnet 5.5 soon. https://t.co/33F4XrBwyT

[Y Combinator @ycombinator]

Congrats to @CalebPeffer, @ericciarla, @nickscamara_, and @firecrawl on their $75M Series B!

Firecrawl turns the web into clean, structured data for AI agents, with more than 1.5 million people building on their API.

They're also launching Alexandria, a knowledge library where agents can search the live web, official data providers, and dedicated indexes of scientific papers, developer docs, and government records. Providers get paid when agents use their data, and Firecrawl plans to open that up to anyone with knowledge worth sharing.

https://t.co/sScCgFgvkZ

[Jay @jayair]

After working on it for a year, I can finally see the end state of the harness

On the backend, it's an execution environment that gives the model maximum degrees of freedom and on the frontend it's a free-form canvas that the model can use to communicate with the user

[SpaceX Super Intelligence @SpaceXSI]

Grok 4.7 is here.

It's a notable improvement over Grok 4.6 at the same price and speed.

[Krzysztof Staroń @KrzysztofStaron]

I love pstack so much, shoutout to @poteto

had a bug for 9 months in one of my projects,
and decided to revisit and throw potato-mode at it.

All it took was proper veryfication, and after 4hours, boom!!! fixed. I've been dealing with it for 9 months, and just like that it's fixed. I'm speechless.

[Jay Lee @joiedevivre01]

@videoclaw Videoclaw

[Jan-Niklas Wortmann @wordmandotdev]

New podcast episode with @dillon_mulroy.

We talked about how he works now, and about what that has done to him.
A lot of uncomfortable truths but I appreciate him sharing his perspective!

0:00 First time the model wrote it the way he would
9:05 He still reads every line
16:43 The work got more exhausting, not less
23:28 The Cloudflare layoff, from the inside
46:26 The tools he actually ships with
48:52 Why GitHub review breaks on AI diffs
54:51 His specs look like TypeScript
56:01 Agents are bad at design and abstractions
1:00:02 Context steering

[Jay Lee @joiedevivre01]

Enjamb

[Rayan Mubarak @rayannmubarak]

Enjamb (YC P26) is the first AI workforce built to work across the full drug program. Biopharma’s hardest work is spread across scientific systems, clinical platforms, regulatory documents, quality records, and years of institutional knowledge. We bring that work together.

[klöss @kloss_xyz]

had Grok @Bot study 300+ GitHub repos of skills

here are the 12 that influenced my setup:

/humanizer by @blader: make your writing and rewrites sound human
https://t.co/YEhDiFZMaR

/unslop by @poteto: removes common AI slop patterns from your writing
https://t.co/mh1i25eckL

/no-ai-slop by @petergyang: make your writing more precise and clearly human
https://t.co/Vgm1xPmw8l

/agent-skills by @addyosmani: agents that interview you before they invent any specs
https://t.co/Y6IYOOcPOB

/skills by @mattpocockuk: a suite of S tier engineering and productivity skills
https://t.co/vk176Bt0hN

/skills by @MengTo: prompt S tier design and UIs like a design system
https://t.co/lZWcrSQklq

/skills by @emilkowalski: design and motion skills that survive real audits
https://t.co/Wqgu8zQoAT

/cartographer by @KingBootoshi: a repo map before your agent wanders
https://t.co/BKpep9ZkTm

/superpowers by Jesse Vincent: stop guessing when you debug, agentic skills framework + complete software dev methodology forcing agents to follow professional workflows
https://t.co/ULxREJMy71

/gstack by @garrytan: memory improvements + spec, review, and QA playbooks so agents cannot silently edit
https://t.co/zCVt87cGiJ

/impeccable by @pbakaus: detect, fix, and recheck with 61 rules
https://t.co/zbMaoYochR

/last30days by @mvanhorn: deep research what people said about a topic over the last month
https://t.co/cUJAXKOpSc

save these and remix them with your Bot!

[Jay Lee @joiedevivre01]

@0xNDX @bot Update: I rebooted my Mac, reinstalled the Grok Bot app, and signed out and signed back in, and it finally worked.

[Jay Lee @joiedevivre01]

@bot I am having critical issues:
- Couldn't reach Grok Bot's computer.
- I tried the recovery to no avail.
- Cannot load a bot from a shared bot (template) link.

[Jay Lee @joiedevivre01]

@0xNDX @bot No worries. Thanks for the heads-up anyway.