Hi, I'm Shriverdhan
Data Scientist & Gen AI Developer
M.Tech Data Science student at NIT Jamshedpur. I design and build intelligent systems, specialized in Large Language Models (LLMs), NLP pipelines, conversational interfaces, and high-performance backend engines.
About Me
Education
National Institute of Technology, Jamshedpur
Master of Technology (M.Tech) in Data Science
Engaged in advanced coursework covering machine learning paradigms, big data architectures, neural networks, and mathematical foundations of data science.
Gen AI & NLP Passion
Architecting Conversational Intelligence
Fascinated by generative models, agentic systems, and retrieval-augmented generation. Focused on translating state-of-the-art AI research into highly performant production systems.
Technical Expertise
Generative AI & LLMs
- LangChain Advanced
- RAG Architectures Expert
- Google GenAI / Gemini API Advanced
- Vector Indexes (FAISS) Advanced
- Prompt Engineering Expert
- LLM Serving (vLLM / Ollama) Intermediate
NLP & Chatbots
- Transformers (HuggingFace) Advanced
- Conversational Agents Expert
- Text Embeddings Advanced
- Sentence Similarity Expert
- NLTK / SpaCy Advanced
- Context Routing Intermediate
Backend & DevOps
- Python Expert
- FastAPI / Flask Advanced
- JWT Security & Auth Advanced
- Docker & Containers Intermediate
- SQL & Database Design Intermediate
- Git / GitHub CI Advanced
Featured Projects
Multi-Document RAG Engine
An end-to-end, security-first Retrieval-Augmented Generation (RAG) system engineered to parse, extract, and answer complex queries from multiple document uploads. Implemented JWT web tokens for access control, semantic context lookup using Facebook AI Similarity Search (FAISS), and high-fidelity text embeddings powered by Google Generative AI.
JWT Authentication & User Service
Designed a modular microservice implementing standard token-based OAuth2 authentication. Featuring secure bcrypt password hashing, token refreshes, endpoint protection, and a highly customizable user database interface.
Optimized LLM Serving Container
A setup configuration to serve large language models locally or in a cloud instance, utilizing optimized Hugging Face pipeline setups, memory pruning techniques, and containerized Docker configs for easy deployment.
Chat with my AI Clone
Shriverdhan Clone
Powered by Client-Side LLM Mock
Quick Queries
Connect With Me
Let's Collaborate!
I'm currently looking for internships, research roles, and full-time opportunities in Data Science, NLP, and Generative AI. Feel free to reach out!