Highly accomplished Founding Chief AI Architect and seasoned Quantitative Researcher with 20+ years of experience in advanced ML system design, quantitative trading, and applied mathematics. Specializing in architecting governed, production-grade ML systems like Mathematically Governed Perception (MGP) for industrial computer vision and radiation-hardened AI runtimes. Proficient in PyTorch/JAX, uncertainty calibration, market microstructure, and latency-aware execution, consistently delivering high-impact, outcome-based solutions with clear IP boundaries. Seeking Staff/Principal ML Architect, Applied Research Scientist, or Quant Research Architect roles.
Remote
→
Summary
As Founding Chief AI Architect at HarmonicQ, Sean Simms led the development and deployment of cutting-edge, production-grade ML systems, including the Mathematically Governed Perception (MGP) platform, driving innovation in industrial computer vision and robust AI architectures.
Highlights
Architected and shipped the Mathematically Governed Perception (MGP) platform, an innovative constraint-aware computer vision system for AEC/industrial drawings, achieving ±2–3% linear-feet accuracy in pilot projects and generating comprehensive Proof Bundles.
Developed and deployed robust evaluation harnesses with calibrated metrics and acceptance tests, ensuring model reliability across on-prem, air-gapped, and cloud VPC environments.
Spearheaded critical reliability efforts, including FaultSynth fault injection, ECC policy management, and runtime guardrails, significantly reducing Silent Data Corruption (SDC) with bounded latency overhead.
Productized the 'Quant/AI Pod' engagement model, streamlining the delivery of CI’d code, model cards, and clear Background-IP separation within 6-8 week build cycles following 10-day discovery phases.
Remote
→
Summary
As a Freelance Quant Researcher/Developer & Consultant, Sean Simms delivered high-impact quantitative solutions across crypto market-making, portfolio engineering, and derivatives coaching for diverse clients.
Highlights
Engineered crypto market-making strategies using HMM, Hawkes processes, and Reinforcement Learning, achieving Sharpe ratios of ≈4.5 with less than 5% drawdown in research simulations.
Developed asymmetric market-making systems for Binance, leveraging multi-modal ML signals and impact-aware execution to manage dynamic risk effectively.
Implemented advanced portfolio engineering techniques, including Bayesian optimization, Mean-CVaR, and Hierarchical Risk Parity, optimizing portfolio construction and PnL under FX/IR volatility.
Provided expert sales & trading coaching on non-linear IR derivatives (IRS, gamma/vol, exotics), covering the full trade lifecycle from inception to settlement.
Remote/Delaware, US
→
Summary
As Quantitative Researcher/Developer at Quantum Signal Trading, Sean Simms developed advanced research and backtesting frameworks, prototyping ML-driven alpha strategies and optimizing execution for North American equities.
Highlights
Developed a comprehensive research and backtesting framework for North American equities, incorporating adaptive VWAP/TWAP and microstructure-aware slippage models to enhance trading strategy performance.
Prototyped and integrated ML-driven alpha strategies (HMM/RL variants), stabilizing portfolio construction and optimizing performance under live market constraints.
Contributed to the design of an advanced Execution Management System (EMS), implementing proprietary routing logic to minimize market impact and improve trade efficiency.
Princeton, NJ, US
→
Summary
As Senior Quantitative Trader at FMI Technologies, Sean Simms designed and implemented minimal-impact execution strategies and developed probabilistic slippage models to optimize trading performance.
Highlights
Designed and implemented minimal-impact execution algorithms, including VWAP/TWAP and hybrid strategies with real-time adjustments, optimizing trade performance in dynamic markets.
Developed sophisticated probabilistic slippage estimation models based on order-book dynamics, directly integrating insights into routing and placement decisions to enhance execution efficiency.
Built and utilized an event-driven backtesting framework to rigorously validate complex trading strategies under realistic market frictions, ensuring robust performance.
Mumbai, India
→
Summary
As Quantitative Consultant for a Private Family Office, Sean Simms engineered statistical arbitrage models and regime detectors, enhancing portfolio timing and resilience.
Highlights
Engineered advanced statistical arbitrage models and regime detectors, significantly improving timing and enhancing portfolio resilience against market shifts.
Kelowna, BC, Canada
→
Summary
As Quantitative Trader at 7Cheetahs, Sean Simms conducted futures research and backtests, yielding measurable uplift in risk-adjusted returns through model refinements.
Highlights
Conducted extensive futures research and backtesting, implementing model refinements that led to a measurable uplift in risk-adjusted returns.
Chicago, IL, US
→
Summary
As a General Surgery Resident at Northwestern University, Sean Simms completed PGY-1 rotations, contributing to protocol-driven quality improvement and research support within a high-stakes environment.
Highlights
Completed intensive PGY-1 rotations across trauma, acute care, and ICU, demonstrating rapid learning and high-pressure decision-making capabilities.
Contributed to protocol-driven quality improvement initiatives and research support, applying rigorous analytical and systematic approaches to complex medical challenges.
Chicago, IL, US
→
Summary
As a Researcher at Shirley Ryan AbilityLab, Sean Simms conducted MEG time-frequency analysis, utilizing advanced machine learning techniques to improve movement-type classification accuracy.
Highlights
Performed advanced MEG time-frequency analysis using VAR/HMM/Bayesian classifiers for movement-type classification, enhancing model accuracy through sophisticated feature engineering.
Applied rigorous scientific methodology and statistical analysis to research, contributing to improved understanding of neurological data.
Chicago, IL, US
→
Summary
As an Independent Futures Trader, Sean Simms developed and executed energy-futures strategies, demonstrating disciplined risk management and independent market analysis.
Highlights
Developed and actively traded energy-futures strategies, consistently applying disciplined risk protocols and independent market analysis to manage positions.
Managed personal capital through speculative trading, gaining practical experience in market dynamics, execution, and risk mitigation.
Published by
Zenodo
Summary
Preprint (under revision) detailing an elliptic PDE formulation with convergence guarantees, highly relevant to physics-informed ML and scientific modeling applications.
ML architecture (CV, sequence, graph), Uncertainty & abstention, Constraint/rule engines, Reliability/safety (fault injection, ECC policy), Evaluation harnesses & CI, Data contracts & model cards, On-prem/air-gapped & cloud VPC deployments, PyTorch, JAX, TensorFlow, scikit-learn, XGBoost, LightGBM, ONNX, CUDA (C++17 basics), FastAPI, gRPC, Docker, Kubernetes (K8s), Airflow, Prefect, GitHub Actions.
Quant research & execution (stat-arb, VWAP/TWAP, slippage/TCA), Market microstructure, Latency-aware execution, HMM/RL variants, Bayesian optimization, Mean-CVaR, Hierarchical Risk Parity (HRP), Non-linear IR derivatives (IRS, gamma/vol, exotics), Bloomberg, Polygon, CCXT.
Python (NumPy, Pandas, Polars), Postgres, ClickHouse, Linux, Grafana, Prometheus.