Ryan J Hamby

I make AI compute fast, efficient, and reliable, from the serving layer down to the hardware and the power behind it.

Software Development Engineer II at AWS, New York. At work: fault-tolerant, multi-tenant payment and IoT systems. Outside work: the layer AI runs on.

Resume GitHub LinkedIn Substack Email

Selected work

OPEN SOURCE · MERGEDvLLM and Firecracker
Root-caused a multimodal crash in vLLM's Qwen3-Omni processing; fixed a VMM panic on ACPI device restore in Firecracker.
RUST · ARROW · 2026FlowState
Rust temporal join engine for point-in-time ML feature pipelines, batch and streaming, with zero-copy Arrow.
GO · GRPC · EARLY BUILDGPU Flight Recorder
Names the straggling rank in distributed GPU training by joining NCCL collective timings with per-GPU telemetry. No validation results yet.
C++20 · 2025–2026Low-Latency Order Book Engine
Lock-free SPSC ingestion and thread-local slab pools, benchmarked on EC2.

Writing

All posts · Subscribe on Substack

GPU Flight Recorder

Mar 2026 GitHub repository Finds the straggler in a distributed data-parallel training job and explains why it is slow. One slow rank makes every other rank wait at every collective; the symptom (step time went up) says nothing about the cause. Flight Recorder joins per-rank collective timing from the PyTorch NCCL flight recorder with per-GPU NVML telemetry, names the straggling rank, and ranks likely causes (thermal or power throttle, ECC errors, degraded PCIe/NVLink, GPU contention, host stall) with the evidence behind each. ...

FlowState

Mar 2026 - Oct 2026 GitHub repository Rust-accelerated temporal join engine for point-in-time ML feature pipelines: join trades, quotes, and signals without look-ahead bias, in batch or streaming. Rust as-of join engine with O(n+m) merge-scan kernels, zero-copy Arrow via the PyCapsule interface, and lock-free UnsafeCell writes across Rayon workers Streaming incremental join with watermark-based emission, per-symbol state, and configurable lateness tolerance; streaming output verified identical to batch 132 Rust tests, including property-based checks of every kernel against reference implementations

Low-Latency Order Book Engine

Sep 2025 - Sep 2026 GitHub repository Minimal, high-performance order book engine in C++ designed to explore the foundations of low-latency trading systems. Includes nanosecond-level order matcher skeleton, lock-free queues, thread-local memory pools, and automated benchmarking on AWS EC2 spot instances. Built C++20 order matching engine with lock-free SPSC ingestion, thread-local slab pools, and inlined matching. On an isolated EC2 c6i.large core over 1M orders: P50 0.21 µs, P99.9 3.1–3.2 µs, 1.4–1.8M orders/sec Automated EC2 spot benchmarking with cloud-init for reproducible, low-cost latency profiling Latency and throughput figures are from the repo README (EC2 c6i.large, isolated core, 3 runs), not a pow run capture. ...

Intelligent Stock Screener

Nov 2025 - May 2026 GitHub repository Fully automated stock screening system scanning 3,800+ US stocks daily using Mark Minervini’s Trend Template methodology. Identifies high-conviction buy/sell signals via phase-based trend classification, relative strength momentum, and fundamental quality filters. Automated daily scans of 3,800+ stocks with 74% API reduction via Git-based fundamental caching and GitHub Actions automation Phase-based classification identifying Phase 2 uptrend breakouts passing 7 of 8 Minervini criteria (50>150>200 SMA alignment, relative strength ≥70, volume confirmation) Risk-managed position management with max 10% risk, min 2:1 R:R ratio, automated stop-loss trailing, and tax-aware filtering for long-term positions Figures on this project are from the repo README, not a captured run. ...

Covariance-Based Macro Trading System

Aug 2023 - May 2026 GitHub repository Systematic ES futures trading system powered by eigendecomposition of economic indicator covariance matrices. Decomposes 8×8 macro factor covariance to identify uncertainty regimes and size positions via macro surprise exposure. Backtested on real FRED data over 17 years (2007–2024): 0.28 Sharpe, 2.10% annualized return, 3.2% max drawdown with half-Kelly, regime-adjusted position sizing C++ factor decomposition engine (Eigen 3.4.0) processing 8 FRED indicators + VIX with AWS Lambda daily automation End-to-end pipeline: FRED API → monthly frequency alignment (downsampling/Hermite interpolation) → spectral decomposition (Σ = UΛUᵀ) → S3, covered by 240+ unit tests Core Hypothesis Markets respond to unexpected economic information, not levels. Raw covariance is misleading—it captures expected relationships. ...

Selected work#

Writing#

Selected work

Writing