Skip to content
View DecodingAzero's full-sized avatar

Block or report DecodingAzero

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
DecodingAzero/README.md

💫 About Me:

Hey, it's Maharnav,

I build systems at the intersection of Machine Learning, Natural Language Processing, Computer Vision, and Large Language Models.

Over the last few years, I have worked on projects involving:

• Fine-tuning Llama 3 (8B) using QLoRA + PEFT
• Building Hybrid RAG pipelines with FAISS + Gemini API
• Designing semantic search engines for healthcare information retrieval
• Developing AI-based computer vision systems for flood depth estimation and urban monitoring
• Working with low-resource NLP systems including language modeling and text generation

Areas I work in

LLM Engineering
Retrieval Systems (RAG / Vector Databases)
NLP Systems Design
Applied Computer Vision
AI Research Engineering

🌐 Socials:

email

💻 Tech Stack:

Python C JavaScript Java MySQL MongoDB Canva Adobe Lightroom Adobe Keras Matplotlib mlflow NumPy Pandas PyTorch scikit-learn Scipy TensorFlow Git GitHub Notion Meta Epic Games Ubisoft Steam nVIDIA Unreal Engine Riot Games OpenCV

📊 GitHub Stats:



✍️ Random Dev Quote

Pinned Loading

  1. dual-architecture-culinary-ai dual-architecture-culinary-ai Public

    Dual-architecture AI recommendation system combining fine-tuned Llama 3 (QLoRA) and Hybrid RAG with FAISS + Gemini to generate context-aware, hallucination-resistant culinary recommendations.

    Jupyter Notebook 1

  2. clinical-semantic-search-faiss clinical-semantic-search-faiss Public

    Semantic search engine for medical literature using MiniLM embeddings, FAISS vector indexing, and dense retrieval for fast, hallucination-free clinical information retrieval.

    Jupyter Notebook 1