Tools, tactics, and taste for product builders.
Join 80,000+ builders from companies like Vercel, HuggingFace, GitLab, and leading startups.
We'll share your email with Substack
Everything else
Jobs every engineer should automate with agents
How PostHog hands off dev chores to scheduled agents – validating fixes in prod, flaky tests, event tracking, and more – with prompts you can copy
Oct 26, 2026 · Jina Yoon · 9 min read
38 things we've learned about building AI products
Lessons from 4 years of building AI products at PostHog: what to build, agent harnesses, MCP, skills, context, evals, costs, and AI UX
Oct 12, 2026 · Ian Vanagas · 8 min read
We're building multiplayer AI. Here's what we've learned so far
Non-obvious lessons for multi-human agent systems: why you need shared context, where collaboration actually happens, and how teams scope work
Sep 28, 2026 · Jina Yoon · 11 min read
What happens to engineers when AI writes all the code?
Agents now open 70% of PostHog's monorepo PRs
Sep 14, 2026 · Ian Vanagas · 9 min read
Read this before deleting your AGENTS.md
The three best ways to keep your AI agent's context healthy as models get better: run /doctor, test your context like code, and ask agents for feedback
Aug 31, 2026 · Jina Yoon · 8 min read
You need to find product-market fit again (sorry)
A case study in disrupting yourself – nine lessons from building PostHog Desktop, our pre-product-market fit bet against our own flagship product
Aug 25, 2026 · Cleo Lant · 7 min read
This post will save you tokens
How to cut LLM and agent token spend without making your product worse: tracking costs, AGENTS.md, MCP audits, prompt caching, and model choice
Aug 18, 2026 · Ian Vanagas · 11 min read
Can software factories actually work?
Software factories let agents ship code with little human review
Aug 11, 2026 · Jina Yoon · 6 min read
What nobody tells you about writing agent skills
Everything we've learned about writing great agent skills at PostHog – progressive disclosure, avoiding over-specification, preventing skill rot, and knowing when not to write one
Aug 3, 2026 · Ian Vanagas · 10 min read
How much can you delegate to agents?
A simple guide to agent autonomy
Jul 27, 2026 · Jina Yoon · 7 min read
2030-shaped software
By 2030, AI agents will be the primary users of software
Jul 20, 2026 · Ian Vanagas · 7 min read
Stop being the code review bottleneck
AI writes code faster than you can review
Jul 9, 2026 · Jina Yoon · 10 min read
We'll share your email with Substack