Professional Work

• Co-founded the Americas-wide AI initiative and now co-lead work across 4 operating centers and 8 engineering functions, scoping and piloting AI workflows for document-heavy and data-heavy EPC tasks.
• Lead the initiative’s use-case pipeline, including idea intake, feasibility review, MVP scoping, project prioritization, stakeholder alignment, and coordination with engineering and digital teams.
• Help translate engineering pain points into practical AI workflow pilots, with emphasis on structured retrieval, vendor-document review, bid tabulation, engineering document intelligence, and review-intensive EPC workflows.
• Lead development of a Python LLM-assisted natural-language-to-SQL pipeline across 3 engineering datasets, turning engineering questions into executable queries for structured retrieval.
• Coordinate technical development of an LLM-assisted bid tabulation workflow that parses large specification packages and normalizes requirement-level information, reducing manual review effort by ~80%.
• Guide technical direction, internal presentations, documentation, pilot framing, and general initiative operations so promising AI ideas can move from scattered concepts into reviewable project workstreams.

• Review vendor instrumentation documentation against P&IDs, PFDs, and Cause & Effect diagrams for blue hydrogen project deliverables, identifying discrepancies and supporting issue resolution.
• Support brownfield and carbon-capture I&C updates by reconciling legacy logic, control narratives, P&IDs, and Cause & Effect documentation within traceability and management-of-change workflows.
• Apply ISA/IEC instrumentation standards, internal specifications, document version control, and QA practices while onboarding across EPC deliverable workflows.

• Established a first-principles molecular modeling capability for carbon-capture R&D, developing reproducible MD/DFT pipelines across HPC and cloud that reduced simulation wall time by ~2/3 and lowered cloud spend.
• Integrated atomistic outputs into diffusion and CFD models through a nucleation-delay reformulation, improving continuum-model accuracy and helping verify the diffusion-scale codebase.
• Evaluated multiple force-field families and benchmarked interfacial energies, diffusion rates, and crystallization barriers against experiments and literature; also designed and supervised crystal-growth experiments.
• Built physics-informed data and ML workflows for automated parameter sweeps, experimental data ingest, terabyte-scale image QA, surrogate-assisted calibration, and surrogate-guided optimization of simulation and experimental design choices.
• Led uncertainty quantification and reproducibility work, authored 15 internal technical notes, mentored teammates, and turned results into funding decks and a 70-page market analysis.

• Rigor & Reproducibility: uncertainty quantification, seeded runs, CI-checked scripts, baselines, and small test cases for fast iteration.
• Reviewability: tools should produce outputs that engineers can inspect, challenge, trace, and hand off.
• Bridging Domains: connect atomistic modeling, continuum assumptions, instrumentation logic, EPC documents, and AI workflows without hiding the assumptions.
• Automation: pipelines that keep analyses current as new images, tables, vendor documents, or project data arrive.
• Communication: concise memos, plots with units, and enough context for non-specialists to disagree productively.

Projects linking physical modeling, embedded systems, and control: vision-guided robot, microcontroller-based controllers, small-scale energy experiments. Teaching assistantship in circuit design; prior aerospace/robotics lab experience on quadcopter stacks and nanosatellite plasma work.