Aditya Goel

I’m a final-year Computer Science & Engineering student with hands-on experience building AI-powered applications using LLMs, Retrieval-Augmented Generation (RAG), and agentic AI design patterns. I enjoy turning complex enterprise problems into reliable, end-to-end GenAI systems with measurable outcomes.

In my Research Intern role at Samsung PRISM (March 2025 – March 2026), I built an AI-powered analytical system processing 100+ product listings and cut manual review time by 60%. I engineered a two-stage agentic LLM pipeline for automated summarization and ranked insight generation across 3+ enterprise use cases.

I’ve also strengthened model performance with retrieval and evaluation: I integrated RAG using Sentence-BERT and FAISS, improving output relevance by 40% over baseline. Using BERTScore on real-world datasets, I achieved 85%+ consistency and implemented safety guardrails to support reliable model behavior.

Outside research, I built an Insider Threat Detection System using Isolation Forest and delivered an executive-facing dashboard with automated alerts—reducing threat investigation time by 50%. I bring a practical, product-minded approach from prototyping to deployment, backed by Python/FastAPI and full-stack experience with React and Spring Boot.