Skip to content
View DevAtomicRelease's full-sized avatar
⚛️
⚛️

Achievements

Achievement: Pair Extraordinaire

Achievements

Achievement: Pair Extraordinaire

Block or report DevAtomicRelease

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
DevAtomicRelease/README.md

Hi 👋, I'm Vishal Sharma

AI Engineer | Architecting Deterministic AI Systems & Agentic Workflows

LinkedIn Email

🚀 Core Focus & Engineering

  • Agentic Workflows & RAG: Building stateful intelligence engines (like Expense AI) using LangGraph and ChromaDB for long-horizon memory retrieval and preventing LLM hallucinations.
  • ML & Vision Pipelines: Deploying real-time MTCNN/InceptionResnetV1 systems and building custom DAPT/TAPT fine-tuning pipelines for BERT/DeBERTa on noisy datasets.
  • Full-Stack AI Architecture: Bridging the gap between ML models and user interfaces by serving highly optimized FastAPI microservices into scalable Next.js and React applications.
  • Advanced Data Modeling: Structuring complex relationships using relational (PostgreSQL), document (MongoDB), and graph databases (Neo4j).

💻 Technical Stack

AI, ML & Data
PyTorch Hugging Face LangChain ChromaDB

Backend & Databases
Python FastAPI PostgreSQL MongoDB

Popular repositories Loading

  1. Real-Time-Face-Recognition-using-Embeddings Real-Time-Face-Recognition-using-Embeddings Public

    This project implements a **real-time facial recognition system** using deep learning–based embeddings for accurate and efficient face identification and verification. It leverages a **pre-trained …

    Jupyter Notebook 1

  2. Local-Document-Q-A-with-FastAPI-and-RAG Local-Document-Q-A-with-FastAPI-and-RAG Public

    A RAG (Retrieval-Augmented Generation) backend built with FastAPI and ChromaDB. It ingests multi-modal documents (including OCR for images) and uses LM Studio to serve a local embedding model and L…

    Python 1

  3. Conversational-AI-Agent Conversational-AI-Agent Public

    Python 1

  4. real-time-face-recognition-hf-demo real-time-face-recognition-hf-demo Public

    JavaScript 1

  5. On-Capture-Image-Quality-Detection-OCR On-Capture-Image-Quality-Detection-OCR Public

    Python 1

  6. Resume-Screener Resume-Screener Public

    Python 1