Released the Kotlin Benchmark for AI coding agents on real-world Kotlin tasks

Released significant update for JetBrains IDEs with Codex agent, MCP management, and approval controls

JetBrains ran a proper A/B benchmark on the Caveman skill for Claude Code, finding 8.5% token savings instead of the claimed 65%.

Released Spec Kit, an open-source toolkit for specification-first AI coding workflows.

Open-sourced Mellum2, a 12B MoE model with 2.5B active parameters for code and reasoning

GitHub launches Spec-Kit, an open-source toolkit for generating technical specifications from natural language.

GitHub Secret Scanning MCP Server reached GA

Released the 'Caveman' tool, claiming it can reduce AI inference costs by up to 75%.

Open-source 'Claude Code' development setup replicating Anthropic engineer's workflow published on GitHub.