7e38e46e65
Latent bug surfaced byf11bab542test audit. `h_mag_per_bucket_kernel` launched with block_dim=32 (single warp) but BUCKET_DIMS terciles are [43, 43, 42] — channels 32..bdim were silently dropped from the sum, producing under-counted Controller D dead-bucket signal in production (perception.rs::h_mag_cfg). Fix: grow block_dim to 128 (smallest pow2 ≥ MAX_BUCKET_DIM=96), expand shared mem from `__shared__ float sdata[32]` to `sdata[128]`, change reduction stride from 16→32→64 (single-warp shuffle equivalent) to 64→32→16→8→4→2→1 (multi-warp block-tree). No atomicAdd, no warp divergence in tail lanes (uniform predicate tid < bdim). Changes: - bucket_transition_kernels.cu: kernel block-tree reduction over 128 lanes, shared mem 128 floats, launch comment updated - perception.rs::h_mag_cfg: block_dim 32→128, shared 32*4→128*4 - bucket_transition_kernels.rs test: launch config matches new contract + bug-surfaced comment replaced with fix-landed reference Test: tests/bucket_transition_kernels::h_mag_per_bucket_kernel_computes _mean_abs_per_bucket now passes (was deliberately failing inf11bab542to surface this bug per feedback_no_todo_fixme — tests assert invariants not observed behavior). Co-Authored-By: Claude Opus 4.7 <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.