Battery / ALD background, 4+ years of software engineering, now trying to bring the two back together.
Google Scholar · ORCID · GitHub
I started in electrochemical materials research, moved into software engineering, and am focus on materials science / chemical engineering.
- Earlier research: Zn-ion batteries, ALD-coated metal anodes, aqueous electrolytes, cycling behavior, and XAS-supported interface analysis
- Software work: 4+ years as an SDE at Amazon, mostly around telemetry, alarms, monitoring, automated mitigation, and internal AI tool
- Current side projects: battery literature mining, small-data ML for ALD/Zn-battery papers, and tools for reading papers without losing the trail
Star badges update automatically.
| Project | Stars | Why it is here |
|---|---|---|
| paper-pilot |  |
A paper-reading tool I built because keeping track of papers gets messy fast |
| JitHubV2 |  |
A Windows GitHub client |
| academia-ml |  |
Small experiments around academic text, paper structure, and research-data cleanup |
| machine-learning-ald-zinc-battery |  |
Small-data ML around ALD-coated Zn-anode papers; more about leakage checks than flashy accuracy |
| battery-paper-crawler |  |
A conservative crawler for battery/electrochemistry papers, with logs so runs are inspectable |
| wis-zn-electrolyte |  |
Notes and workflow pieces around Zn-battery electrolyte/interface reading |
Battery interfaces and diagnostics
I care about battery data that still points back to the cell: cycling behavior, degradation signals, electrolyte/interface changes, and the boring-but-important details that make experiments comparable.
ALD and thin-film processing
My ALD/Zn-anode work made me interested in process-property questions: when a coating actually changes interfacial chemistry, electrochemical stability, or device behavior.
Materials informatics
I am cautious about generic "AI for materials" claims. I am more interested in data quality, uncertainty, literature-derived datasets, and models that still have a physical interpretation.
- Scientific Python: pandas, NumPy, scikit-learn, notebooks, data-cleaning pipelines
- Research tooling: literature harvesting, metadata extraction, run logs
- Software engineering: TypeScript, C#, backend/frontend tooling, production monitoring
- Scientific communication: LaTeX, paper summaries, structured research notes