Uses

A breakdown of the tools, languages, frameworks, and platforms I use to build AI systems, machine learning pipelines, and full-stack applications.

AI & Machine Learning

LangChain and LangGraph are my go-to frameworks for building agentic workflows and multi-agent AI systems. LangGraph is especially powerful for orchestrating stateful agent pipelines.
OpenAI API powers most of my LLM and embedding workflows. I use the embeddings API extensively for semantic search and RAG pipelines paired with Pinecone as my vector database of choice.
TensorFlow and Scikit-learn for model training, evaluation, and classical ML workflows. I also use Hugging Face Transformers for NLP tasks like zero-shot classification and sentiment analysis.
Pandas and NumPy for data manipulation, with Matplotlib and Seaborn for visualization and exploratory data analysis.

Cursor is my primary editor for AI-assisted development. I also reach for VS Code and PyCharm depending on the project.
FastAPI is my backend framework of choice for building AI-powered APIs. It pairs naturally with async LangGraph agents and Python ML stacks.
Next.js and React for full-stack web applications. Most of my deployed AI projects use this stack with Vercel for hosting.
Google Colab and Jupyter Notebook for exploratory data analysis, model experimentation, and ML prototyping.
Supabase and Firebase for databases and auth in full-stack projects, alongside Clerk for authentication in Next.js applications.

Languages	Python, JavaScript, TypeScript, Java, SQL
AI / ML	LangGraph, LangChain, OpenAI API, TensorFlow, Scikit-learn, Transformers
Vector / Data	Pinecone, Pandas, NumPy, BigQuery ML
Frontend	React, Next.js, MUI, Gradio
Backend	FastAPI, SpringBoot, Node.js
Databases	Supabase, Firebase, MongoDB
Cloud / Deploy	Vercel, Cloudflare, Google Cloud
Tools	Cursor, VS Code, PyCharm, Jupyter, Google Colab, Git, Figma
Certification	eCornell Machine Learning Foundations (August 2025)