Projects
Things I've built & explored
🤖 Agentic AI
Job Buddy — Autonomous Job Search Agent
A multi-phase autonomous job search agent that parses GitHub READMEs to build a vectorized knowledge base of personal projects, scores job postings against my resume using Google Gemini, and autonomously extracts job descriptions from external career sites via Tavily AI.
Multi-phase autonomous agent with RAG and Gmail integration
MLOps
LLM Buddy - LLM Evaluation & Benchmarking Platform
LLM Buddy is an automated benchmarking pipeline that evaluates multiple LLMs across coding, reasoning, summarization, and RAG tasks using LLM-as-a-judge scoring. Buddy uses LLaMA 4 Maverick as an independent judge to score responses on accuracy, clarity, and completenessand tracks results in MLflow for experiment reproducibility and persist results to GCP BigQuery for long-term trend analysis. The solution also has automated weekly runs via GitHub Actions with GCP Workload Identity Federation and Live leaderboard deployed on GCP Cloud Run.
Automated LLM evaluation pipeline with live GCP leaderboard
🧠 AI Dev Tool
LeetCode Buddy — AI Study Tool
A intelligent LeetCode buddy to make you a Leetcode Master. Buddy has a plan with 170+ problems across 3 phases and buddy also helps you track your solutions and patterns covered for easy revision. The project allows you to submit the leetcode solutions, uses Gemini AI to enrich your submission with metadata and commits them directly to GitHub. Built on Netlify serverless functions, the app also classifies each solution by pattern, generate approach summaries, infer time/space complexity, surface key insights and gotchas, and auto-flag problems for revisit based on self-rating.
AI-powered leetcode study tracker with automatic GitHub commits
Multimodal ML · NLP + Vision
Multimodal Sentiment Classifier
Built a multimodal sentiment classifier on the MVSA dataset using cross-modal attention and gated fusion. Finetuned RoBERTa for text and ResNet-50 for images, fused with a learned gating mechanism.
Val Macro-F1: 0.63 on MVSA dataset
