Code, decisions, evidence
Applied ML, tool-calling agents, and data products. Explore the implementation, what it demonstrates, and where its limits are.
8 projects
Ask a cricket question and inspect the tools behind the answer. Gemini selects endpoints for player statistics, matchups, match analysis, and charts.
Every tool call exposes its arguments and structured result beside the answer.
Turn synthetic health journals into traceable symptom signals, trends, and visit summaries. Every extracted signal retains its source text and evidence span.
A source-first pipeline from journal entry to an immutable, reviewable report.
Practice interviews with resume-specific questions, gap detection, and adaptive follow-ups using retrieval and a fine-tuned language model.
Resume-aware interviews with 18 gap types and inspectable, persistent session state.
Explore natural-language sports queries, structured intent, compiled SQL, and chart recommendations in a working sample-data demo.
Inspect the query plan and SQL alongside the result, with explicit sample-data mode.
Extract job descriptions and recruiter emails into validated records with local NLP, duplicate checks, and optional Notion synchronization.
Local extraction first; an optional Gemini fallback fills incomplete fields.
Compare statistical, neural, and tree-based forecasts of hourly electricity demand across 368 clients grouped into four load-profile clusters.
A chronological 2014 holdout after training on 2011-2013 observations.
A local workflow for assistant-authored resume tailoring, rendered-PDF validation, application tracking, and optional Notion synchronization.
Publish a local PDF only after structural, page-count, and keyword checks pass.
Explore suspicious network activity through ML predictions, a threat dashboard, and a conversational interface for analyst questions.
A hackathon prototype connecting behavioral features with an analyst-facing workflow.