business-sim-openenv / inference.py

Commit History

  1. Victory Lap Strategy: Maximizing final spikes and adding success badge

    ychanakya12
    710a337
  2. Final Charge Protocol: Maximum late-game aggression to spike graphs

    ychanakya12
    b282012
  3. Cash Reserve Protocol: Implemented budget floor and 80% success threshold

    ychanakya12
    24ad1dd
  4. Legacy CEO Update: Client Acquisition and Brand Equity tracking

    ychanakya12
    59940af
  5. Hyper-Growth Protocol: Enforcing high-efficiency profit hunting

    ychanakya12
    29fcdaf
  6. Ironclad CEO: Fixed task confusion and implemented 40% budget safety cap

    ychanakya12
    0b538e5
  7. Lean-to-Giant strategy: Quarter 1 burn reduction and late-game expansion

    ychanakya12
    348bdfe
  8. Exponential Growth strategy: Phase-based investment and profit spikes

    ychanakya12
    3f6ce4d
  9. Strategic Stabilization: Balancing skill retention and aggressive growth

    ychanakya12
    366df60

  1. Optimize Agent strategy (ROI analysis + Salary Optimization) for 0.5+ scores across all tasks

    ychanakya12
    ac9dd65

  1. MODEL UPGRADE: switching to openai/gpt-oss-120b for optimized benchmark scores

    ychanakya12
    ac06697
  2. STRICT OPENENV LOG FORMAT: key=value pairs, success field, no pipes

    ychanakya12
    947e6ce
  3. GRANULARITY FIX: extremely granular graders, ASCII-only logs, 3-decimal [END] lines

    ychanakya12
    31f0e47
  4. Complete clean rewrite of inference.py - fixes corrupted indentation, guaranteed score in (0.01,0.99)

    ychanakya12
    5844f89
  5. Guarantee exactly 3 [END] lines with score strictly in (0.01, 0.99) — catch all exceptions

    ychanakya12
    819f3dc

  1. Clamp score in inference.py [END] line and all graders - strictly (0.01, 0.99)

    ychanakya12
    185b1f3
  2. Added aggressive profit/risk guardrail to baseline agent

    ychanakya12
    c074bf7
  3. Final Hackathon Submission updates

    ychanakya12
    8f447ab