Spencer
Purdy

Spencer Purdy
MACHINE LEARNING ENGINEER

I build production-grade ML systems at the intersection of quantitative finance and deep learning, from Physics-Informed Neural Networks for options pricing to Deep RL agents for market making and end-to-end MLOps pipelines.

What I build

I'm a Machine Learning Engineer specializing in production ML systems and applied deep learning. I graduated Magna Cum Laude from Auburn University with a BS in Computer Science and an undergraduate certificate in AI Engineering.

My work spans the full ML lifecycle, from model architecture and training to deployment, drift detection, and monitoring. I build systems designed to hold up in production, not just in a notebook.

ML & AI

Fine-tuning (LoRA/PEFT) RAG Systems Neural Networks Ensemble Methods Reinforcement Learning Physics-Informed ML

MLOps & Production

Model Versioning Drift Detection A/B Testing Monitoring Hyperparameter Tuning

Tools & Development

Python PyTorch scikit-learn XGBoost Optuna ChromaDB Transformers Git

Selected work

01
Deep Learning · Physics-Informed ML · Derivatives Pricing

Neural Network Options Pricer
with Greeks & Volatility Surface

Physics-Informed Neural Network (PINN) with a [128, 256, 256, 128] Tanh architecture that enforces the Black-Scholes PDE directly in the training loss via automatic differentiation. Trained on real AAPL options data from yfinance with a curriculum that ramps the physics weight from 0 to 0.01 over training. Computes all five Greeks (Delta, Gamma, Vega, Theta, Rho) via PyTorch autograd using stock-price-augmented training data. Prices American options via Least Squares Monte Carlo with Laguerre polynomial basis functions across 10,000 simulation paths, and uses a PPO agent for dynamic delta hedging. Includes arbitrage detection across put-call parity and butterfly spread convexity checks.

PyTorch · PINNs · PPO · Autograd · LSM · Stable-Baselines3
02
Reinforcement Learning · Market Microstructure · HFT

Deep RL Market Maker
with Order Book Dynamics

Custom Gymnasium environment with 12-dimensional observations and a 2-D continuous action space for bid/ask offset placement. PPO agent trained for 500,000 timesteps using an Avellaneda-Stoikov inspired reward: per-step PnL change plus a quadratic inventory penalty and a terminal liquidation penalty. Models order flow with a Poisson arrival process (λ = 10 orders/second), VPIN-based toxicity tracking, and a square-root market impact function with adverse selection. Hard inventory limit of 1,000 shares enforced at execution. Transaction costs: 1 bps fee plus 0.5 bps slippage. Policy and value networks: MLP (64→64) with Tanh. Expected benchmarks: Sharpe 0.5–2.0, fill rate 40–80%, spread capture 30–60%, max drawdown <5% of initial capital.

PPO · Gymnasium · PyTorch · Stable-Baselines3 · Reward Shaping
03
MLOps · Production ML · Model Monitoring

Automated MLOps Framework
for Customer Churn Prediction

End-to-end MLOps platform built on the IBM Telco Customer Churn dataset (7,043 customers, 20 features, ~26% churn rate). Trains an ensemble of XGBoost, LightGBM, and Random Forest models with Optuna hyperparameter optimization (30 trials, 5-fold cross-validation) and SMOTE for class balancing. Achieves ~0.85 ROC-AUC and ~80% accuracy on a held-out 20% test set, with inference latency under 100ms per prediction. Features a SQLite model registry with versioning, Kolmogorov-Smirnov drift detection (p < 0.05 threshold), A/B testing at 95% confidence, and SHAP-based feature explainability.

XGBoost · LightGBM · Optuna · SHAP · KS Drift Detection
04
Portfolio Optimization · Applied ML · Quantitative Finance

AI-Powered Multi-Asset
Portfolio Optimizer with Risk Management

Production-grade portfolio optimization across a 10-asset universe (SPY, QQQ, IWM, EFA, EEM, TLT, IEF, GLD, SLV, VNQ) combining Modern Portfolio Theory, Black-Litterman Bayesian posterior estimation, and ML-driven return forecasting. XGBoost model trained on 100+ engineered features (rolling returns, volatility, moving averages, RSI, volume) with an 80/20 train/test split. LSTM: 2-layer, 128-unit recurrent network on 60-day input sequences, trained with MSE loss and Adam optimizer. ML-Enhanced mode blends 70% historical signal with 30% ML predictions. Fama-French 5-factor model (Market, SMB, HML, RMW, CMA) for risk decomposition. FinBERT sentiment analysis on a 30-day Finnhub news window. Risk constraints via VaR/CVaR at 95% confidence; Monte Carlo stress testing with up to 50,000 paths over 5-year horizons. Rebalancing triggered at 5% weight drift with 10 bps transaction costs. Models pre-trained on H100 GPU.

PyTorch · XGBoost · CVXPY · FinBERT · LSTM · Fama-French

Where I've worked

Jun – Jul 2024
Software Developer Intern
CGI
  • Built full-stack Manager Dashboard using .NET 8 MVC for employee promotion and PTO tracking
  • Led 8-person development team as Tech Lead, driving sprint completion and project deliverables
  • Designed and implemented RESTful APIs with Entity Framework for scalable data management
May – Aug 2023
Computer Systems Engineering Intern
General Dynamics Mission Systems
  • Optimized mission-critical server performance on Littoral Combat Ships through rigorous testing
  • Diagnosed and repaired enterprise hardware including MFDs and UPS using military procedures
  • Validated network infrastructure through OS configuration and NTP synchronization testing
Oct 2022 – Dec 2024
Software Quality Assurance Analyst
Auburn University Office of Information Technology
  • Executed software testing protocols ensuring compliance with functional requirements
  • Managed defect tracking and resolution workflows using Jira and qTest for development teams
  • Collaborated with cross-functional teams to identify, document, and resolve software issues

Background

Georgia Institute of Technology
Master of Science in Computer Science
Specialization in Machine Learning
Aug 2026 – Aug 2028
Machine
Learning
Specialization
Aug
2028
Expected
Auburn University
Bachelor of Science in Computer Science
Undergraduate Certificate in AI Engineering
Aug 2021 – May 2025
3.73
GPA / 4.00
Magna
Cum Laude
Honors
Relevant Coursework:
Machine Learning Artificial Intelligence Data Mining Multi-Agent Systems Software Modeling & Design Software Construction Introduction to Algorithms Discrete Structures Database Systems I Statistics for Engineers Linear Algebra Calculus III

Let's talk.

Open to full-time Machine Learning Engineer roles.

spencer@spencercpurdy.com