PhD-turned-founder: LLMs flipped which criteria kill programming tool startups
jimmykoppel · x · 2026-10-03
Jimmy Koppel shares the framework he used throughout his PhD to evaluate programming tool startup ideas. A good tool startup must solve a problem that is: hard enough to need a tool, easy enough that you can build it, general enough for a large market, and legible enough that buyers know the pain exists.
- Before LLMs: the middle two criteria ruled out most ideas — his PhD focused on building smart multi-language tools without wading through 500-page language specs.
- After LLMs: the first and fourth criteria are now the bottleneck. Many tools compete with Claude skills or home-grown vibed apps, and agents get stuck in loops that neither users nor VPs of engineering can even detect.
The core shift: LLMs collapsed the cost of building, moving startup viability hinges to whether the problem is genuinely hard and whether the value is legible to buyers.
More from coding & agent
- New Claude Code plugin renders pasted images as thumbnails above the prompt — jarrodwatts · 2026-10-03
- Coinbase for Agents adds bracket, stop-limit and TWAP orders plus automated feedback loop — kleffew94 · 2026-10-03
- PhantomEnvironments: 7B LLM Trained in Synthetic RL Environments Matches Agents 10x Its Size — CShorten30 · 2026-10-03
- OpenAI introduces dot: cross-app memory, context and Codex task coordination — OpenAIDevs · 2026-10-03
- MIT Interactive Diagrams: From Attention to Mixtral and DeepSeek-V3 Architectures — vtabbott_ · 2026-10-03
- Agent Aware Architecture: an open spec to make sites walkable for AI agents — sierracatalina · 2026-10-03