|
I'm an AI Engineer and Computer Science graduate with 2+ years of experience building AI agents, generative AI applications, and end-to-end ML/DL systems. I specialize in RAG pipelines, multi-agent workflows (LangGraph/LangChain), and LLM evaluation, with a strong foundation in computer vision and taking models from notebook to production.
|
|
AI Agents & Automation
Generative AI & LLMs
ML & Deep Learning
Computer Vision
Backend, Deployment & Tools
| Project | Problem | Approach | Result |
|---|---|---|---|
| DataScribe — AI-Powered Multi-Agent Data Analysis Assistant | Manual data analysis is slow and error-prone — inspecting data, writing code, validating results, and building reports all by hand. | LangGraph multi-agent workflow (planner, programmer, critic, reporting) with LLM-driven planning, Pandas analysis, retry logic, structured outputs, and LangSmith tracing/evaluation on a FastAPI backend. | End-to-end assistant generating executable analysis workflows and reports, evaluated via accuracy, node-level metrics, LLM-judge scoring, and P50/P99 latency. |
| CareerPrep_AI — AI-Powered Interview & Career Assistant | Interview prep is fragmented across separate tools for research, question generation, mock interviews, and feedback. | LangGraph/LangChain orchestration across planning, research, mock interviewing, and feedback stages, with LLM reasoning, retrieval/web research, and evaluation-oriented tracing. | Integrated career assistant coordinating multiple interview-prep tasks through one orchestrated workflow. |
| GlowGenie — AI-Powered Skin Disease Detection & Recommendation System (FYP) | Skin-condition screening is hard to access quickly, especially where dermatology specialists are limited. | YOLOv12-based object detection for skin conditions, fused with symptom-based ML probabilities via late fusion; evaluated on precision, recall, IoU, and mAP. | YOLOv12s achieved ~0.666 precision, 0.581 recall, and 0.63 mAP@0.5 — a multimodal screening workflow combining image and symptom evidence. |
- Artificial Intelligence (ML, Deep Learning, Communication) — NAVTTC
- IBM AI Engineering Specialization; Python for Data Analysis (Pandas & NumPy) — Coursera
- Neural Networks and Deep Learning; Improving Deep Neural Networks (Hyperparameter Tuning, Regularization & Optimization) — Coursera