Poolside open-sources its eval data as its CEO says coding is the path to AGI
eliebakouch · x · 2026-07-24
Poolside leans into openness with public eval data and a model factory
A post praising Poolside highlights two things the company has been unusually open about: it shipped a small coding model that reportedly beat Thinky Machines at coding, and it published its full evaluation dataset.
The linked podcast quote says Poolside's CEO thinks coding is the path to AGI, that its Model Factory runs 10,000–20,000 experiments a month, and that new models can be turned around in as little as eight weeks. It also argues that persistence and verification may matter more than raw intelligence, and that agents are already helping build the next generation of models.
The attached benchmark UI screenshot shows a broader evaluation archive with multiple public benchmarks, runs, and cost/step tracking, underscoring the company's emphasis on measurable coding performance rather than vague capability claims.
Related event: Poolside Details Coding as a Path to AGI(2 posts)→
More from coding & agent
- Opinion: AI Agent Harnesses Will Evolve From Products to Libraries — samgoodwin89 · 2026-07-24
- An engineering lead uses OpenLoomi to sync GitHub, Linear, and PR reviews — Yuuyake · 2026-07-24
- Independent search lifts Fable, Sol, Grok, and Gemini accuracy in real-world tasks — ycombinator · 2026-07-24
- Topview Launches Marketing MCP Integrating E-commerce Data and Content Generation — azed_ai · 2026-07-24
- VideoTreeSearch: Organizing Videos as Trees for Grounded Long Video QA — mohitban47 · 2026-07-24
- A new AI agent checklist moves execution authority out of the model — Jay299792458 · 2026-07-24