reefContinual-learning infra for self-improving agents: serve traffic (OpenAI/Anthropic-compatible), match feedback to runs, train weights via Slime/SGLang or evolve the harness, ship versioned updates.
Why switchBoth implement a self-improving loop that evolves an agent's harness and weights; SIA is the paper's Meta/Target/Feedback agent recipe, Reef is production infra with serving, feedback matching and versioned delivery around it.
Full comparison → evoKarpathy-style autoresearch on any codebase: /evo:discover instruments the benchmark, /evo:optimize runs tree search with parallel subagents in worktrees. Plugin for Claude Code, Codex & co.
Why switchBoth run autonomous improve-against-a-benchmark loops with paper-adjacent rigor. evo points coding agents at YOUR codebase (tree search over commits); SIA generates a task-specific agent and evolves the agent itself — harness and weights.
Full comparison → JITJIT-Agent: a meta-agent that writes a task-specific harness on the fly — memory, planning, action and capability modules emitted as code — and revises it from traces at test time.
Why switchBoth come from the 'improve the harness, not just the model' line of work. SIA's Meta/Target/Feedback loop evolves a task agent's harness and weights over time; JIT-Agent generates a fresh harness per task at test time and keeps the generator frozen.
Full comparison → rrsiGoogle Research's Regularized Recursive Self-Improvement: evolves an agent harness (prompts, tools, memory, sub-agents) with constraints on the search so gains hold out of distribution.
Why switchBoth run a meta-loop that evolves a task agent's harness against a benchmark; SIA also trains weights, RRSI keeps the model frozen and regularizes the search so gains survive out of distribution.
Full comparison → autoresearchKarpathy's autoresearch loop as an installable skill for Claude Code, OpenCode and Codex: constraint + mechanical metric + autonomous modify→verify→keep/discard iteration.
Why switchSame lineage — the autoresearch loop: metric, modify, verify, keep or discard. The skill hill-climbs your code with a fixed agent; SIA makes the agent itself the thing that improves, up to and including its weights.
Full comparison →