Stories by scosman

Animating SVGs with LLMs

RAG Isn't One-Size-Fits-All - Here's how to Tune It

Vercel.com: Do not use Vercel

Show HN: Synthetic data generation for evaluating RAGs

Pelicans Riding Bicycles

Mail your parents a tailscale node

Show HN: Breaking IDE silos with MCP for prompts and dev-tools

Show HN: SecretShare – Easy, secure one time secret sharing CLI

Eval-maxing an AI FFmpeg command generator

Show HN: Give coding agents MCP access to lint/test/format with 1 YAML file

Show HN: Kiln – AI Boilerplate with Evals, Fine-Tuning, Synthetic Data, and Git

I wrote 2000 LLM test cases so you don't have to: LLM feature compatibility grid

Ask HN: LLM for Browser Inspector/DevTools/DOM?

Many small evals beat one big eval, for AI Products

When Fine-Tuning Makes Sense: A Developer's Guide

Show HN: Kiln – GitHub+Jira+Figma for AI product development

Using Computer Fans for Home Air Movement

A prompt engineer's guide to fine-tuning

Show HN: Build your own AI evals in a minutes – benchmark to human prefs

Ask HN: What does "Used by engineers at [Company]" for OSS mean? Stars/emails?

1
2