Skip to content
View solankinitish's full-sized avatar
2 followers · 10 following

Block or report solankinitish

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
solankinitish/README.md

Hi, I'm Nitish Solanki 👋

Self-directed developer based in Delhi, building toward a remote GenAI engineering internship.

I build LLM systems from first principles — no tutorials, no copy-paste. Every component understood before it's written.

What I Work With

LLM ENGINEERING   → LangChain, Groq, Ollama, Mistral, Llama 3.1
VECTOR DBs        → Pinecone, ChromaDB
EMBEDDINGS        → SentenceTransformers
ML                → scikit-learn, scipy, HuggingFace Transformers, distilBERT
BACKEND           → FastAPI, Uvicorn, Pydantic
FRONTEND          → Streamlit
DEVOPS            → Docker, Google Cloud Run, GCS
MATHS             → Linear Algebra (MIT 18.06), Multivariable Calculus, Optimization
DEEP LEARNING     → Neural networks, Backpropagation, CNNs, RNNs (Keras, TensorFlow)
LANGUAGE          → Python 3.11

Currently

  • Building production-style GenAI systems — RAG pipelines, stateful memory, ML-grounded LLM orchestration
  • Studying LLM internals, RAG evaluation, and prompt engineering
  • Trying to use Mathematical and Classical ML based heuristic lens in this big-models world

👉 Explore the implementations of these concepts in my pinned repositories below.

Looking For

Seeking Applied AI / LLM Systems Engineering Internships focused on retrieval, orchestration, and AI infrastructure

📧 nitishsolanki888@gmail.com

Twitter LinkedIn

Pinned Loading

  1. know-thyself know-thyself Public

    A Stateful AI orchestration system that combines persistent memory, classical ML grounding, and LLM-based reasoning to deliver context-aware long-term interactions across multiple domains

    Python 1

  2. knowledge-ops knowledge-ops Public

    Web-grounded RAG research engine — live search, multi-hop query decomposition, hybrid answerability reranking, and local LLM inference

    Python 2

  3. structured-text-engine structured-text-engine Public

    Fully local RAG backend — semantic retrieval, cosine reranking, and grounded LLM inference over a persistent vector store

    Python 1