product idea: If you are interviewing for a new role, upload the JD, company, any other context, and have AI generate a koans like study guide for you. Can vary depending on how well the language or framework for which you will be assessed

What are people using for workflow management these days in Rails? I use @sidekiq but I really miss something like @temporalio. I'm not sure I want to stand up that infra right now. Is there a happy medium?

The hardest part about AI is getting a human to review my massive vibe coded PRs

AI coding with @cursor_ai @claudeai @GoogleAI etc has made me a 10x engineer. I produce 10x the technical debt 10x faster.

Mmmm thanks SF I love the smell of shit and weed in the morning.

The rotisserie chicken at Costco has thicker skin than @garrytan

Founders, If your software engineers were anti AI would you fire them?

San Francisco is not ready for RTO. @SFBART has been working on this escalator for at least a month. Today more escalators out at Embarcadero. Hard to believe we call SF a tier 1 city.

Super cool LLM visualization of inference bbycroft.net/llm

Farmer Poll 🧑‍🌾 If presented the opportunity for tech interests to buy your farmland for $250,000 per acre, but would be *required to quit farming forever*, would you sell?

The @linear bullet list to subtasks may be one of my favorite product features ever.

Someone needs to make a special rewards card/wallet for AI spend specifically. It would be a good distribution network for new products

I'm so tempted to get a group of people together building 1-person startups using AI agents to help them build. Weekly meetings. Daily standups. Sharing tips, ideas and pitfalls. It'd be free but you'd have to apply. If that sounds interesting, reply and we'll see what happens.

holy shit

remember when Deep Seek news hit and everyone freaked out that demand for GPUs might go down? I'm using more and more AI on more powerful and powerful models and I don't see that stopping anytime soon. Also #PeaceNotWar

Who is working on a compressed response schema for Gen Ai models? We are running into token limit output issues with JSON schema. I need the model to give me gzipped JSON 😂

What is the cheapest model to handle natural language logic? Let’s say I’m building an email marketing system and it has automations. One tactic could be to respond with a follow up email if the user seems interested by replying to the email message. Is there a cheap model that can handle “is the response to this email following the instructions ‘please reply with your favorite color to receive the promo code’”? Writing the code for this would be insane and would never catch all edge cases. But would be a perfect fit for an LLM

I just got laid off from Block today. It was a solid 2 and half years. I'll be looking for my next role immediately, if anyone has any openings or know of any places where I might be a good fit lmk. I'm interested in roles around: - dev rel - dev education - dev experience