Skip to content
View aniketpoojari's full-sized avatar
1 follower · 4 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report aniketpoojari

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
aniketpoojari/README.md

Hi, I'm Aniket Poojari 👋

AI/ML Engineer | Agentic Systems | MLOps

I specialize in building autonomous AI agents, fine-tuning LLMs, and architecting scalable MLOps pipelines. I'm passionate about creating systems that don't just generate text, but reason, reflect, and verify their own outputs.

🚀 Featured Deployed Projects

🧠 Agentic AI Research Assistant

Autonomous Researcher with Self-Reflection.

  • Built with LangGraph to implement a Generate-Critique-Refine loop.
  • Reduces LLM hallucinations by fact-checking claims against real-time web results.
  • Tech: Python LangGraph FastAPI

🎭 BiLoRA: Dual Adapter AI Assistant

Multi-task Fine-Tuning on a Single Backbone.

  • Fine-tuned Microsoft Phi-3 using dual LoRA adapters for specialized Code and Docstring generation.
  • Dynamic adapter switching at inference time.
  • Tech: PyTorch DVC MLflow

📄 OmniDoc QA — Multi-Modal RAG

Intelligent Document Q&A System.

  • Processes PDFs and Websites with structural and table extraction.
  • Built-in monitoring dashboard for latency, cost, and retrieval quality.
  • Tech: LangChain Groq Docker

🌸 Text-to-Image Generation

Latent Diffusion Model from Scratch.

  • Distributed training on AWS SageMaker with DeepSpeed.
  • Optimized for inference using ONNX and TensorRT.
  • Tech: PyTorch AWS ONNX

✍️ Technical Writing (Medium)

I write about the engineering behind my projects. Check out my latest blogs:

🛠️ Tech Stack

Languages: Python SQL AI/ML: PyTorch LangChain HuggingFace MLOps: DVC MLflow Airflow Cloud/Infra: AWS Docker FastAPI ONNX

📬 Connect with me: LinkedIn | Portfolio

Pinned Loading

  1. LLM-Eval-Platform LLM-Eval-Platform Public

    Full-stack platform to evaluate, red-team, and A/B test LLM applications with 216 adversarial attacks and statistical rigor.

    Python

  2. Enterprise-AI-Assistant-MCP Enterprise-AI-Assistant-MCP Public

    Production-grade Multi-Modal RAG system for intelligent document Q&A with structural extraction and real-time observability.

    Python

  3. Agentic-AI-Research-Assistant Agentic-AI-Research-Assistant Public

    Autonomous AI research agent using LangGraph to eliminate LLM hallucinations via a Generate-Critique-Refine self-reflection loop.

    Python 1

  4. OmniDoc-QA-LLM-RAG OmniDoc-QA-LLM-RAG Public

    Agentic NL-to-SQL assistant using Model Context Protocol (MCP) with multi-layer guardrails and token cost tracking.

    Python

  5. Fine-Tuning-LLM Fine-Tuning-LLM Public

    Multi-task fine-tuning system using dual LoRA adapters on Microsoft Phi-3-mini for specialized code and docstring generation.

    Python

  6. Text-To-Image-Diffusion Text-To-Image-Diffusion Public

    End-to-end latent diffusion pipeline featuring distributed training on AWS SageMaker, DeepSpeed optimization, and ONNX/TensorRT inference.

    Jupyter Notebook