9951b8b8cb
Replace the serial curiosity_fwd_bwd_per_block kernel (CUR_TOTAL_PARAMS=11306 loop iterations × block-level shared-memory reduction per-iteration = 1085ms) with a cuBLAS GEMM pipeline matching the existing curiosity inference path: Forward: curiosity_prepare_input → GEMM1(W1) → bias_leaky_relu → GEMM2(W2) → mse_fwd_grad Backward: gemm_dw(dW2) → bias_grad_reduce(db2) → gemm_dx(d_hidden) → leaky_relu_bwd → gemm_dw(dW1) → bias_grad_reduce(db1) New CUDA kernels added to curiosity_training_kernel.cu: - curiosity_mse_fwd_grad: +b2 in-place, d_pred = 2/CUR_OUTPUT*(pred-target) - curiosity_leaky_relu_bwd: gates d_hidden by sign of post-activation hidden - curiosity_bias_grad_reduce: sum dy[N, D] over batch → grad_b[D] GpuCuriosityTrainer rewritten with CuriosityGemm (dedicated cuBLAS+cublasLt handle) + intermediate buffers (input_buf, hidden_buf, pred_buf, d_hidden_buf). Reuses forward kernels from curiosity_inference_kernel.cu. Keeps curiosity_adam_step. Drops partial_grads buffer (max_blocks*11306 floats saved). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Foxhunt
Production HFT trading system in Rust.
Architecture
The workspace contains 32 crates organized as follows:
Core Libraries (16)
| Crate | Purpose |
|---|---|
trading_engine |
Order processing, FIX 4.4, IB TWS, SIMD, RDTSC timing |
risk |
VaR, Kelly, circuit breakers, kill switches, compliance |
risk-data |
Risk data types and shared structures |
trading-data |
Trading data types |
ml |
DQN Rainbow, PPO, TFT, Mamba2, ensemble inference |
ml-data |
ML data types and feature definitions |
data |
Market data ingestion and storage |
backtesting |
Replay engine, strategy tester |
adaptive-strategy |
Ensemble execution, microstructure analysis |
common |
Shared types, resilience, error handling |
storage |
S3 and local model storage |
model_loader |
Model serialization and loading |
market-data |
Market data feed handlers |
database |
PostgreSQL access layer (SQLx) |
config |
Configuration management |
tli |
CLI commands and tooling |
Services (8)
| Service | Purpose |
|---|---|
backtesting_service |
gRPC backtesting service |
broker_gateway_service |
FIX routing, broker connectivity |
trading_service |
Core trading operations |
ml_training_service |
Model training orchestration |
data_acquisition_service |
Market data acquisition |
trading_agent_service |
Autonomous trading agents |
api_gateway |
gRPC API gateway with auth |
web-gateway |
Axum REST + WebSocket gateway |
Frontend
web-dashboard/ -- React 19 + TypeScript + Vite + TradingView charts.
Building
# Check compilation (no PostgreSQL required)
SQLX_OFFLINE=true cargo check --workspace
# Run tests for a specific crate
SQLX_OFFLINE=true cargo test -p <crate> --lib
# Clippy
SQLX_OFFLINE=true cargo clippy --workspace
ML Models
Four production model architectures on Candle v0.9.1 with CUDA:
- DQN Rainbow -- Deep Q-Network with prioritized replay, dueling heads, noisy nets
- PPO -- Proximal Policy Optimization with GAE and LSTM policies
- TFT -- Temporal Fusion Transformer for multi-horizon forecasting
- Mamba2 -- State space model for sequence prediction
Each model has a standalone trainer and a UnifiedTrainable adapter for the hyperopt pipeline.
Infrastructure
- Git: Gitea at
git.fxhnt.ai(Tailscale-only), Scaleway DEV1-S - Observability: OpenTelemetry OTLP (env
OTEL_EXPORTER_OTLP_ENDPOINT) - Database: PostgreSQL with SQLx offline mode for CI
License
Proprietary. All rights reserved.