Projects
Projects I want this site to point to first.
These are the projects that best capture how I like to work: optimization-heavy ideas, decision
tools, and applied machine learning systems that try to be genuinely useful.
01
Full Stack ML pipeline for hotel reservation prediction
This project uses the Python data science stack, Docker, Jenkins, and MLflow to predict the
likelihood that a hotel reservation will be canceled.
Why it matters: it shows how modeling work becomes more useful when it is wrapped in a clean,
reproducible deployment pipeline instead of living only as an experiment.
Stack and idea: end-to-end ML operations, production-minded data workflows, and cloud-ready
deployment thinking.
Open project repository
02
Stochastic optimization for energy arbitrage against renewable assets
An advanced Pyomo model for minimizing risk while buying and selling power in an energy market
using clearing rules and renewable asset production scenarios.
Why it matters: this is exactly the kind of decision problem where optimization, uncertainty, and
energy economics collide in a useful way.
Stack and idea: Pyomo, stochastic optimization, scenario modeling, and market-facing decision
support for renewable-heavy systems.
Open project repository
03
Apps, experiments, and technical products
Small applications that package analysis, automation, or machine learning into something easier to
use, test, and share.
Why it matters: building software is how I pressure-test ideas. Shipping even a compact tool
reveals what is useful, what is confusing, and what deserves to become a deeper project.
Typical ingredients: product thinking, fast iteration, pragmatic engineering, and enough design
to make the tool approachable.