Skip to content
View paramramit305-a11y's full-sized avatar

Block or report paramramit305-a11y

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
paramramit305-a11y/README.md

Amit Parmar

AI/ML engineer in training — BSc IT student (Gokul Global University) building GenAI and agentic systems end to end, from model logic to deployment.

📍 Palanpur, Gujarat · Open to AI/ML internships (Ahmedabad or remote) · paramramit305@gmail.com


Currently

  • Completed a self-correcting RAG system (CRAG) with LangGraph — per-document grading, query reformulation, live web search fallback, deployed on Streamlit
  • Going deeper on agentic systems — tool-using agents, persistent memory, multi-agent coordination (Agno framework)
  • Applying for my first AI/ML internship

Stack

Languages: Python

GenAI & Agentic: LangGraph LangChain Agno ChromaDB Tavily PyTorch

LLM APIs: Groq OpenAI compatible

Deployment: Flask Streamlit Render HuggingFace


Featured projects

CRAG — Corrective RAG Research Assistant — Self-correcting RAG built with LangGraph: grades each retrieved document individually, reformulates queries on ambiguous results, falls back to live web search when local docs fail. · Live

YouTube Video Analyzer — Agno agent that takes a YouTube link and returns a timestamped breakdown using tool calls, not a single prompt. · Live

AI Personal Assistant — Flask app on Groq's API, running in production on Render. · Live

See all repos →



LinkedIn · HuggingFace

Pinned Loading

  1. agentic-ai-youtube-analyzer agentic-ai-youtube-analyzer Public

    Agno agent that takes a YouTube link and returns a timestamped breakdown of the video, using Groq + YouTubeTools. Built while learning the Agno framework.

    Python 1

  2. crag-research-assistant crag-research-assistant Public

    A self-correcting RAG system built with LangGraph — grades each retrieved document individually, refines relevant knowledge, and falls back to live web search when local documents are insufficient.

    Python 2

  3. RAG-Pipeline RAG-Pipeline Public

    RAG pipeline from scratch — PDF ingestion, local embeddings, ChromaDB vector store, and LLM-agnostic generation (Groq/OpenAI/Anthropic) using LangChain.

    Jupyter Notebook 1

  4. ai-personal-assistant ai-personal-assistant Public

    Flask-based AI Assistant powered by Groq API

    HTML 1

  5. text-summarizer-t5-finetune text-summarizer-t5-finetune Public

    Fine-tuned T5-small on SAMSum dataset for dialogue summarization | HuggingFace Transformers · FastAPI · Gradio · Deployed on HF Spaces

    HTML 1

  6. credit-risk-streamlit-app credit-risk-streamlit-app Public

    💳 End-to-end ML web app for loan approval prediction | Logistic Regression | 88% Accuracy | F1: 80.9% | Deployed on Streamlit

    Jupyter Notebook 4