About
Sagnick Das
AI Engineer and full-stack developer based in Howrah, India. I build systems where machine learning meets production software — and make sure they actually work at scale.
Background
I'm a final-year Computer Science student specializing in Data Science at MCKV Institute of Engineering, with a GPA of 8.41/10. My work sits at the intersection of AI research and software engineering — I don't just build models, I deploy them.
I've shipped production AI pipelines delivering 37% accuracy gains and 75% reduction in processing time. My systems run on AWS ECS with Docker, handling real traffic from real users. I've architected RAG pipelines, built autonomous coding agents, and hardened ML models against adversarial attacks.
On the full-stack side, I've freelanced for multiple clients — delivering production web apps serving 500+ users, building REST APIs handling 1,500+ monthly requests, and automating deployment pipelines that cut content update time from 2 days to 30 minutes.
Skills
Languages
- Python
- JavaScript / TypeScript
- Java
- SQL
- R
- C++
AI / ML
- LLMs & Prompt Engineering
- RAG & Vector Databases
- LangChain, Hugging Face
- TensorFlow, PyTorch
- Scikit-learn
- ChromaDB, Embeddings
Frameworks
- FastAPI, Flask
- React, Next.js
- Express.js, Node.js
- Spring Boot
- TensorFlow Lite
Infrastructure
- Docker
- AWS (ECS, ECR)
- PostgreSQL, MongoDB
- REST API design
- CI/CD, Linux
- DNS, SSL, Cloud
Approach
Research-driven
I start with the problem, not the tool. Whether it's adversarial robustness or semantic search, I dig into the research before writing code.
Production-first
A model that doesn't deploy is a notebook. I design for containers, concurrent pipelines, and cloud infrastructure from day one.
Full ownership
From database schema to DNS configuration, I handle the entire stack. No handoffs, no gaps, no surprises.