Interview Lens
ARI module that turns a candidate's take-home project into a structured, role-specific interview brief.
Portfolio
Filter by area, then open a project to step through how it works.
ARI module that turns a candidate's take-home project into a structured, role-specific interview brief.
Multi-agent tutoring platform that turns uploaded materials into personalized study outputs with teacher oversight.
Capability broker that routes agent tasks to the right tool, redacts sensitive context, and gates paid execution.
Human-approval layer for autonomous coding agents: approve, steer or deny each tool action from a private Telegram chat while execution stays local.
End-to-end TTC reliability platform covering 1 million+ delay events across bus, subway, GTFS, and live-alert data.
Course research on fine-tuning Llama 3.1 8B with graph-augmented prompts for job-shop scheduling (JSSP).
Cut hourly traffic forecast error by 54% with a GRU, then showed the simpler Poisson GLM is the better anomaly alarm (ROC-AUC 0.94 vs 0.71).
ROS2 perception-to-manipulation pipeline using YOLOv8, OpenCV transforms, and ArUco calibration for precise robotic sorting.
Planning assistant to structure exam preparation timelines and task prioritization.
Fine-tuned multilingual transformer summarization and translation pipeline for legal text, packaged as a Flask service.
Compared ML and deep learning models for Uber NYC driver pay, reaching almost 12% lower RMSE than baseline, with Tableau feature analysis.
Regression modeling and feature analysis pipeline for India weather forecasting competition data.
Natural-language interface for uploaded SQL databases, translating prompts into executable SQL queries.
Retrieval-first document QA with LangChain + FAISS and OpenAI Agents SDK workflows, reaching 80% accuracy on a manual test set (N ≤ 50).
Semi-automated grading system with BERT-based NLP and OCR to evaluate typed and handwritten responses.
Android Kotlin app using Reddit APIs and Glide for lightweight meme streaming UX.