jgrusewski d6f0538ac0 docs: CUDA backtest & GPU-resident training design spec ...

Three-phase plan to eliminate all GPU→CPU roundtrips from training:
- Phase 1: Seal training loop (persistent GPU epoch state, async monitoring)
- Phase 2: Vectorized CUDA backtest kernel for hyperopt evaluation
- Phase 3: General-purpose GPU backtester replacing CPU SIMD path

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-11 09:11:27 +01:00

..

infra

feat(infra): GPU training launcher with local/cloud routing

2026-02-23 11:19:58 +01:00

plans

docs: CUDA backtest & GPU-resident training design spec

2026-03-11 09:11:27 +01:00

superpowers

docs: add zero-CPU DQN training hot path implementation plan

2026-03-10 15:49:40 +01:00