Optimization engineer with industry experience developing optimization models for energy applications, including production-grade Unit Commitment and AGC software deployed at utilities worldwide. PhD (Michigan State, 2026) specializing in evolutionary and surrogate-assisted optimization. Strong foundation in MIP/LP, multi-objective optimization, and constraints handling — with hands-on experience in Gurobi, COIN-OR/CBC, and Python scientific stack.
Optimization engineer with industry experience developing optimization models for energy applications, including production-grade Unit Commitment and Automatic Generation Control software deployed at utilities worldwide. PhD from Michigan State University (2026) specializing in evolutionary and surrogate-assisted (machine-learning-assisted) optimization under Prof. Kalyanmoy Deb (creator of NSGA-II/III).
Strong theoretical foundation in constraints handling, multi-objective optimization, Linear Programming, and Mixed-Integer Programming with hands-on experience using Gurobi, COIN-OR/CBC, Pymoo, and the Python scientific stack (NumPy, Pandas, SciPy, Scikit-learn). Prior industry experience at Eaton Corporation in systems modeling and simulation for commercial vehicle powertrain programs.
Seeking Applied Scientist and Optimization Engineer roles where rigorous algorithmic research meets production engineering at scale.
I design and develop large-scale optimization algorithms for power generation and renewable integration within AspenTech's Generation Management System (GMS) platform. My work bridges optimization, operations research, software engineering, and energy systems, with a focus on delivering efficient, production-ready optimization solutions.
Built high-fidelity multi-physics models supporting commercial vehicle development for GM, Ford, Isuzu, DAF, and FCA. Work spanned variable valve actuation strategies (CDA, engine braking, LIVC, EEVO), vehicle transmission sump temperature analysis including electric vehicles, engine simulations, electrical circuit breaker mechanics, and transmission/clutch vibration diagnosis. Resolved torsional vibrations in Eaton Endurant transmissions, preventing fault code generation and warranty claims.
Developed MagneX Gen2 power distribution hardware designs using DFSS (Design for Six Sigma) methodology. Designed core for cold shrink high-voltage cable accessory joining technology and investigated material formulations for electrical and mechanical properties.
First author on all research publications. Work spans surrogate-assisted evolutionary optimization, multi- and many-objective optimization, and power systems. Published in IEEE TEVC (IF 14.3), SWEVO (IF 8.2), Nonlinear Dynamics, GECCO, CEC, EMO, and WCCI. Google Scholar ↗ ORCID ↗
Open-source implementation of the heterogeneous-evaluation-time surrogate-assisted NSGA-III framework. Handles many-objective problems where different objectives carry different computational costs, deciding per-solution which need high-fidelity evaluation. Published in IEEE Transactions on Evolutionary Computation, 2025 (IF 14.3).
Production Security-Constrained Unit Commitment solver deployed at utility clients across North America. Designs and maintains large-scale MIP/LP formulations in Gurobi and COIN-OR/CBC for generation dispatch, reserves, and energy accounting within AspenTech's Generation Management System.
A mixed-fidelity evaluation framework for surrogate-assisted evolutionary multi-objective optimization. Decides per-solution which objectives and constraints need high-fidelity evaluation — based on feasibility, non-domination probability, evaluation cost, and surrogate uncertainty — avoiding wasteful full evaluations. Extends HET-NSGA-III to constrained problems. Published in Swarm and Evolutionary Computation, vol. 100, 2026.
Population-based methods aren't going away. Here's why evolutionary algorithms remain essential for real-world engineering problems — and what they do that gradient descent fundamentally cannot.
The SCUC problem is one of the hardest MIPs run in near-real-time every day by grid operators worldwide. Here's what it actually is and why energy AI needs to understand it.
A practical guide to choosing the right function approximator for expensive optimization problems. When does Kriging beat a neural network? More often than you'd think.
Open to Applied Scientist and Optimization Engineer roles. Also happy to collaborate on research or discuss surrogate-assisted optimization.