CV

• Led a research agenda to discover techniques for LLMs to dramatically improve open-source developer workflows
• Proposed unit-test automation as the research objective, designed evaluation frameworks (hybrid embedding + code-coverage evals) and experiments, and developed a graph-structured reasoning framework for automated test generation across AndroidOS
• Built structured, hierarchical datasets of Android's source code to improve one-shot performance and fine-tuning
• Collaborated with DeepMind and CoreML teams to define data specifications and quality criteria for training Google's internal models on open-source code—the first time these models were trained on open-source codebases
• Developed an unsupervised root-cause analysis method for Android OS infrastructure errors using T5X embeddings and density-based clustering, reducing incident response time by 90%

• Technical lead on API correctness and code-coverage infrastructure for Android Mainline (modularizing the Android operating system)
• Enabled continuous AndroidOS releases across 800+ million devices and 5 platforms
• Collaborated with product managers and UX designers to deliver interactive developer tools used by 200+ Google engineers
• Built interactive tools that let developers identify and resolve coverage gaps, reducing time-to-resolution by 95%

• Optimized Android smart-sync performance, enabling developers to sync to the latest best-known AndroidOS baseline up to 99% (one day) faster
• Developed a photomosaic image-generation algorithm that recreates a source image by tiling thousands of images from a database (the Met gallery database)

• Developed the first presubmit testing infrastructure for the technology stack
• Designed UI features for the BULQ iOS app and web landing page
• Built internal front-end tools that let non-engineering teams independently update their live web content

• Earned High Departmental Honors in Computer Science for a year-long independent research project culminating in a qualifying exam, thesis defense, and public presentation
• Designed a novel interpretable model for general game playing: a two-stage multiagent Monte Carlo search that converges to Nash equilibria and runs in real time without pre-training

• Invented constraint-optimization algorithms that generate digital artworks
• Used acyclicity, connectivity, planarity, and two-colorability as constraints to render images as connected spanning trees, two-color convex tilings, and morphing polygon grids with flexible vertices (documented at madebymath.art)

• Conducted a literature review on the political, economic, and engineering challenges of piezoelectric energy as a viable renewable
• Proposed uses of piezoelectricity on Oberlin's campus, evaluating feasibility and expected energy output across candidate sites

• Mentored and taught 12+ early-career engineers across several teams through Google's Engineering Residency and Early Career Immersion programs
• Guided project scoping, design decisions, and code reviews to accelerate onboarding and independent contribution

• Guided introductory CS students through lab exercises and debugging, reinforcing core programming concepts
• Graded labs and provided detailed feedback to support student learning

• Gained hands-on medical experience in organ donation compatibility testing, processing protein samples, and managing patient data

• Assisted filmmakers in interviewing participants and documenting the Smithsonian Peruvian Folk-Like Festival through film and photography