I build production-ready AI systems, scalable FastAPI services, and agentic LLM pipelines.
- End-to-end multimodal and generative AI systems
- FastAPI backends for high-throughput inference
- LLM reasoning, RAG pipelines, and agentic workflows
- ML lifecycle, evaluation, and model serving
- OCR, vision, video pipelines, embeddings, and vector stores
- Async architectures and orchestration patterns
- Languages: Python, JavaScript, Java, C
- Frameworks: FastAPI, LangChain, Streamlit, Flask
- ML/DL: PyTorch, TensorFlow, scikit-learn, SentenceTransformers
- Vision and Media: PaddleOCR, PyMuPDF, PDFPlumber, OpenCV, FFmpeg
- Databases: MongoDB, Postgres, SQLite
- Infra and Data: Pandas, NumPy, Docker, Git, Linux
- RAG and LLMs: OpenAI API, Ollama, Pinecone, FAISS
- Scraping: Selenium, BeautifulSoup, requests
- Architecting AI systems from ingestion to inference
- Designing RAG and multimodal reasoning workflows
- Building scalable FastAPI microservices
- Production-grade async pipelines
- Converting research prototypes into stable systems
Build usable, reproducible, scalable AI systems with clear abstractions and observability.
|
Collaborative project repository for CO-CODE event. 
|
Project repository for Recursion Hackathon 6.0. 
|
|
Team JBBR LOC7 coding repository. 
|
Data-driven project for JBBR DataQuest. 
|
|
Backend powering JBBR systems and services. 
|
Project for Mumbai Hacks built by JBBR. 
|
