Skip to content
View Saty9956's full-sized avatar
  • Dr B R Ambedkar National Institute of Technology
  • Jalandhar

Block or report Saty9956

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
Saty9956/README.md

Hi there, I'm Satyartha 👋

AI Engineer & Data Scientist

I am currently pursuing my M.Tech in Artificial Intelligence at NIT Jalandhar and working as a Graduate Technical Intern at Intel Corporation, specializing in AI and Systems Engineering. My work focuses on building practical, enterprise-grade AI systems that create measurable impact, spanning Agentic AI, LLM-powered applications, Retrieval-Augmented Generation (RAG), predictive modeling, and applied machine learning.

I enjoy combining strong data foundations with modern AI architectures to solve real-world business problems.

🚀 What I'm Currently Working On

🤖 Agentic & Generative AI

  • Building multi-agent workflows for autonomous task execution and intelligent decision systems.
  • Designing complex RAG pipelines with semantic retrieval, routing, and evaluation layers.
  • Improving LLM reliability through advanced prompt engineering and validation frameworks.
  • Exploring open-weight models and optimizing fast inference systems.

📊 Data Science & Machine Learning

  • Developing predictive models for critical business use cases, such as risk scoring and expected loss.
  • Running A/B testing and statistical experiments to drive data-informed product decisions.
  • Building explainable ML systems using feature importance and interpretability tools (SHAP).
  • Architecting end-to-end ML pipelines from data preprocessing to deployment.

🛠️ Tech Stack

AI / LLM / Agentic Systems: LangChain | Vector DBs (FAISS) | RAG Pipelines | Semantic Routing | Groq API | Prompt Engineering | Multi-Agent Workflows
Data Science / Machine Learning: Python | SQL | Scikit-learn | XGBoost | LightGBM | SciPy | SHAP
Deep Learning & Vision: TensorFlow | Keras | CNNs | Computer Vision Models
Cloud, Backend & Data Analysis: AWS (RDS, S3) | FastAPI | Streamlit | Pandas | NumPy | Matplotlib | Seaborn | Git

📌 Featured Projects

🔹 Omni-Swarm: Agentic Q-Commerce Economics

A Multi-Agent AI system simulating quick-commerce dispatch, autonomous rider negotiation, and real-time unit economics (arbitrage, surges, and cash burn) using Llama 3, AWS (RDS + S3), and Streamlit.

🔹 Agentic Search Reliability Engine

A production-grade Agentic RAG system utilizing semantic routing, hybrid retrieval (FAISS + Pandas), and a built-in audit layer to drastically reduce hallucinations in enterprise search environments.

🔹 Clinical-RAG-Assistant

A high-fidelity Medical AI Assistant built with RAG, FastAPI, and FAISS to provide source-backed clinical information, featuring a zero-latency safety triage layer for reliable healthcare querying.

🔹 Credit Risk & Expected Loss Model

An enterprise-grade Credit Risk pipeline that predicts loan defaults (Probability of Default) using LightGBM and Explainable AI (SHAP), designed for regulatory compliance and decision support.

🔹 A/B Testing Feature Optimization

An end-to-end statistical experimentation pipeline evaluating e-commerce UI changes. It features statistical significance testing, effect size analysis, and localized segment checks to drive product decisions.

🔹 AI MCQ Generator

An automated assessment system that generates high-quality Multiple Choice Questions from uploaded PDF content using the Groq API, Mixtral-8x7B, and a Streamlit frontend.

📫 Connect With Me

⭐ Always open to collaborating on AI, Data Science, and innovative product ideas.

Pinned Loading

  1. Omni-Swarm-Agentic-Q-Commerce Omni-Swarm-Agentic-Q-Commerce Public

    Multi-Agent AI system simulating quick-commerce dispatch, rider decisions, and real-time unit economics using Llama 3, AWS, and Streamlit.

    Python

  2. Clinical-RAG-Assistant Clinical-RAG-Assistant Public

    A high-fidelity Medical AI Assistant using RAG (Retrieval-Augmented Generation), FastAPI, and FAISS to provide source-backed clinical information with a zero-latency safety triage layer.

    Python

  3. Agentic-Search-Reliability-Engine Agentic-Search-Reliability-Engine Public

    Production-grade Agentic RAG with semantic routing, hybrid retrieval (FAISS + Pandas), and audit layer to reduce hallucinations in enterprise search.

    Python

  4. MCQ-Generator MCQ-Generator Public

    AI-powered MCQ generator using Groq, Mixtral-8x7B and Streamlit with PDF input

    Jupyter Notebook 1

  5. Credit-Risk-Expected-Loss-Model Credit-Risk-Expected-Loss-Model Public

    Enterprise-grade Credit Risk & Expected Loss (EL) pipeline predicting loan defaults using LightGBM and Explainable AI (SHAP) for regulatory compliance.

    Python

  6. AB-Testing-Feature-Optimization AB-Testing-Feature-Optimization Public

    End-to-end A/B testing pipeline evaluating e-commerce UI changes. Features statistical significance testing, effect size analysis, and localized segment checks to drive data-informed product decisi…

    Python