Ridges AI launches database engineering agent competition after hitting 76% on linting tasks
bittingthembits · x · 2026-09-10
Ridges AI, the coding-agent competition running on Bittensor's SN62 subnet, continues iterating: earlier competitions covered broad software engineering (repo repair, multi-language problems) and a beta product, Ridgeline, lets users assign software work to agents.
Competition 27 introduced linting tasks, with team-reported results of 76% on a hidden test set within a week and cost down to $0.02 per task. The newly launched Competition 28 targets database engineering — agents compete on fixing bugs, optimizing queries, and working with ORMs and GraphQL. The thesis: inefficient queries are a real pain point where specialist agents can outperform generalists.
More from coding & agent
- Developer Builds a Tiny Dock for Keeping Tabs on Her Swarm of AI Agents — Wattenberger · 2026-09-11
- bury-bench: a deterministic, zero-LLM-judge benchmark scoring coding agents on ADHD-friendly answers — Obluness · 2026-09-11
- Gemini API Docs Move Into AI Studio, with Docs MCP and Agent Skills Support — _philschmid · 2026-09-11
- ZenML built its first RL environment to benchmark coding agents on ML workflow tasks — strickvl · 2026-09-11
- Reviewer: Relume is the first AI-first design tool that doesn't skip the design process — michalmalewicz · 2026-09-11
- OpenAI agents hijacked a German wiki to share notes — dev builds Pod, an MCP for agent tool reviews — saheedniyi_02 · 2026-09-11