02-rl-reward

Build RL reward signals using the OpenJudge framework. Covers choosing between pointwise and pairwise reward strategies based on RL algorithm, task type, and cost; aggregating multi-dimensional pointwise scores into a scalar reward; pairwise tournament reward for GRPO on subjective tasks (net win rate across group rollouts); generating preference pairs for DPO/RLAIF; and normalizing scores for training stability. Use when building reward models, scoring rollouts for GRPO/REINFORCE, generating pr

Install

KitScout test notes

Not tested by KitScout yet. When our harness runs it, pros, cons and the tasks we tried appear here, with the date and evidence.

Similar tools

• REA 456,474 ★
  Reverse engineer anything from your terminal or agent with one CLI and MCP server.
• ask-matt 323,337 ★
  Ask which skill or flow fits your situation. A router over the skills in this repo.
• chief-of-staff 323,337 ★
  Pursue a long-running goal in a single session by co-ordinating subagents.
• claude-handoff 323,337 ★
  Hand the current conversation off to a fresh background agent that picks up the work immediately.
• grill-me 323,337 ★
  A relentless interview to sharpen a plan or design.