I’m a first-year Computer Science major at UCLA interested in building practical software and learning fast. I’m currently exploring full‑stack web development and applied ML/NLP.

• LA Blueprint — PATH (Resource management app | TypeScript/Next.js + Supabase) Shipped production features across tickets, inventory, and user/admin management: built reusable item search with URL-synced filtering, implemented outgoing/incoming ticket workflows with shared components, and helped support admin-facing operations.

• Bloom (React + TypeScript | Vite, Zustand, ReactFlow) Built a client-side “living journal” that turns Markdown headings into interactive mind maps in real time: implemented bi-directional Page↔Grove sync, draggable node layouts, and structured views for outlining and navigation.

• SAF3AI (Python / Transformers / scikit-learn) Fine-tuned encoder/LLM-based classifiers for agentic AI governance and prompt-injection detection: trained DeBERTa and DistilBERT variants on the public deepset/prompt-injections dataset, added custom safe/injection examples, evaluated precision/recall/F1, switched to a multilingual DistilBERT variant after finding German bypass cases, and tuned the decision threshold to improve recall without hurting precision.

• ELL Bot (FastAPI + React | Docker + Azure) Built a chatbot platform to support refugees with a full-stack architecture: developed a FastAPI backend and React frontend, set up CI/CD for frontend deployment from GitHub, and containerized the backend for deployment.

• Languages: Python, C++, TypeScript/JavaScript
• ML / AI: Hugging Face Transformers, PyTorch, scikit-learn, model evaluation
• Web: React, Next.js, Vite
• Backend / Infra: FastAPI, Docker, Supabase, Azure

I’m looking for opportunities to learn from strong engineering teams through internships and mentorship. If you’re building something interesting in web, ML/NLP, or developer tooling, I’d love to connect.

• LinkedIn: https://www.linkedin.com/in/nseerapu/
• GitHub: https://github.com/neeleshseerapu