Skip to content
View JS12540's full-sized avatar
💭
Learning and Building
💭
Learning and Building
28 followers · 165 following

Achievements

Achievement: Pull Sharkx2Achievement: YOLO

Achievements

Achievement: Pull Sharkx2Achievement: YOLO

Block or report JS12540

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
JS12540/README.md

Jay Shah

AI Engineer | LLM Systems | MLOps | Cloud-Native Architectures

About Me

AI Engineer focused on building scalable, production-grade ML systems and LLM-powered applications.
Strong experience across the full model lifecycle — from training and optimization to deployment on cloud-native microservices.

I enjoy solving real-world problems using AI systems that are reliable, measurable, and efficient.

Featured Project: Aarogya — AI Healthcare Assistant (WhatsApp-Based)

Aarogya is an AI-driven healthcare assistant that helps users understand prescriptions and medical reports, discover generic medicine alternatives, and ask follow-up health questions — directly via WhatsApp.

Technical Highlights

  • WhatsApp Business API (WABA): Scalable conversational healthcare workflows
  • LLMs: Simplified prescription & medical report explanations
  • Vision + OCR: Processed scanned prescriptions and lab reports using image-based LLM pipelines
  • Multilingual Translation: Real-time language adaptation for regional accessibility
  • Parallel Processing: Concurrent pipelines for image parsing, report analysis, and contextual Q&A
  • Context Retention: Maintains session memory to support unlimited follow-up queries per report

Impact

  • Reduced confusion around medical prescriptions and lab reports
  • Improved affordability by surfacing generic medicine alternatives
  • Delivered healthcare insights in minutes without requiring app downloads

Try Aarogya on WhatsApp:
https://wa.me/message/OMBNFBED6JCFJ1

Open Source

concall-parser

Financial research utility that extracts management commentary and analyst Q&A from earnings call transcripts.

PyPI: https://pypi.org/project/concall-parser

Technical Stack

Languages

Go · Python · SQL · JavaScript

ML & AI

PyTorch · TensorFlow · scikit-learn · Hugging Face Transformers
LangChain · LangGraph · LlamaIndex
RAG Systems · Vector Search · Prompt Engineering · Model Quantization (ONNX, INT8)

Databases

PostgreSQL · MongoDB · Pinecone · Qdrant

Cloud & DevOps

AWS · GCP · Docker · Kubernetes · CI/CD Pipelines · GitHub Actions

MLOps

MLflow · DVC · Airflow · Model Evaluation Pipelines

Backend & APIs

FastAPI · Flask · REST Microservices · Async Processing

Notable Achievements

  • Kaggle Bronze Medalist (Top 13%) – LEAP Atmospheric Physics Competition
  • Built optimized quantized models with sub-millisecond inference latency
  • Designed high-precision retrieval systems (RAG + semantic caching)
  • Developed production LLM agents for Text2SQL and document retrieval

GitHub Activity (Streaks)

GitHub Streak

Connect

LinkedIn: https://www.linkedin.com/in/jay-shah
Email: jayshah0726@gmail.com

Pinned Loading

  1. Deep-learning-book-projects-and-learnings Deep-learning-book-projects-and-learnings Public

    Jupyter Notebook

  2. Generative-Adversarial-Network-for-Digit-and-Mnist-dataset Generative-Adversarial-Network-for-Digit-and-Mnist-dataset Public

    Jupyter Notebook

  3. Text2ImageSearch Text2ImageSearch Public

    Python

  4. Fine-Tuning-LLM Fine-Tuning-LLM Public

    Jupyter Notebook

  5. Real-Estate-Chatbot Real-Estate-Chatbot Public

    Jupyter Notebook

  6. classification_model_mlops classification_model_mlops Public

    Jupyter Notebook