fc0311de77
Per spec §5.2 and feedback_no_partial_refactor. Adds bucket-routing metadata to the Checkpoint envelope: - Renames tau -> tau_all (same HIDDEN_DIM shape) - Adds bucket_id_per_channel: Vec<u8> - Adds bucket_channel_offset: Vec<u32> - Adds bucket_dim_k: Vec<u32> - Adds heads_w_skip_offset: Vec<u32> heads_w_skip stays at [N_HORIZONS x HIDDEN_DIM]=640 floats; compaction to [HIDDEN_DIM]=128 deferred to Task 10's forward dispatch update. new_random zero-initializes bucket metadata; Phase 1->2 transition (Task 9) populates them. CfcTrunk field `tau_d` renamed to `tau_all_d` across: - crates/ml-alpha/src/cfc/trunk.rs (struct + save_checkpoint + load_checkpoint + new_random + save_load_roundtrip bit-equality test with 4 new u8/u32 assertions on the bucket metadata fields) - crates/ml-alpha/src/trainer/perception.rs (6 self.trunk.tau_d references + 1 trunk.tau_d upload + 1 comment line) ml-backtesting/tests/checkpoint_smoke.rs needed no edit: its only consumer surface is CfcTrunk::load_checkpoint(&dev, &cfg, &ckpt), whose signature is unchanged. The 3-consumer atomic-migration audit is satisfied because consumer #3's API contract is preserved. Greenfield envelope (per existing cfc/trunk.rs:25-28 convention - no backward compat). Verified: - cargo check --workspace --all-targets: clean (pre-existing cupti/insert_batch errors in crates/ml/tests excluded per task scope) - cargo test -p ml-alpha --lib: 33 passed - cargo test -p ml-backtesting --lib: 33 passed - cargo test -p ml-alpha --lib save_load_roundtrip -- --ignored: PASS (bit-equality on tau_all_d + 4 new bucket metadata fields, RTX 3050) - cargo test -p ml-backtesting --test checkpoint_smoke: compiles clean Co-Authored-By: Claude Opus 4.7 (1M context) <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.