Skip to content
View hithesh-mr's full-sized avatar
💭
Here to learn and explore 😎
💭
Here to learn and explore 😎

Block or report hithesh-mr

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
hithesh-mr/README.md

Hi, I'm Hithesh M R

Data Scientist - AI Engineer @ 4flow

GenAI | RAG | Agentic AI | End-to-End ML Systems


Education

  • M.Tech in Computational and Data Science
    NITK Surathkal (2022 – 2024) — Graduated with Distinction
  • B.Tech in Mechanical Engineering
    NIE Mysore (2017 – 2021)

Experience

Data Scientist – AI Engineer

4flow | Nov 2025 – Present | Bengaluru, India

  • Building GenAI-powered business solutions using RAG, LLMs, and agentic workflows
  • Developing scalable AI systems for enterprise automation and analytics
  • Working across cloud platforms, APIs, and production-grade deployments

Associate Data Scientist – AI Engineer

Gramener (A Straive Company) | Sep 2024 – Nov 2025 | Bengaluru, India

PGIM Dealio — SEC EDGAR Analytics Platform

  • Built FastAPI endpoints and Pydantic schemas for SEC filings and derived analytics (HFA, CAP, COMP)
  • Designed a cache-first ingestion system with normalization and quarter/year rollover handling
  • Integrated Azure OpenAI for metric mapping, prompt workflows, and response caching
  • Developed Azure Functions for background jobs with secure authentication and Blob/Table logging
  • Created a lightweight JS + Tailwind UI and generated PDF/Word reports using ReportLab, PyMuPDF, Jinja2

Malaysia Traffic Flow Visualization (Penang Island POC)

  • Built predictive models for traffic speed and delay forecasting
  • Processed geospatial data using Turf.js, Mapbox GL JS, and ArcGIS
  • Developed Flask APIs to serve data to interactive dashboards
  • Enabled real-time visualization of traffic flow and predictions

RAG Framework for Financial Document Querying

  • Implemented pgvector-based vector search with PostgreSQL (3072-d embeddings)
  • Designed fusion ranking methods (RRF, LCRF, CombSUM) for improved retrieval accuracy
  • Built APIs for ingestion, embedding, and query workflows
  • Supported multi-cloud deployment and developer tool integrations

Research Publication

First Author
From Pixels to Prognosis: Exploring from UNet to Segment Anything in Mammogram Image Processing for Tumor Segmentation
IEEE CONIT 2024


Current Focus

  • AI agents for business problem solving
  • Traffic forecasting and time-series systems
  • Financial document search with RAG
  • PostgreSQL (pgvector), OpenAI API, LangChain, cloud deployments

Areas of Expertise

  • GenAI, RAG, Agentic AI systems
  • Vector search and knowledge graphs
  • Geospatial analytics and visualization
  • Medical image segmentation (U-Net, SAM)
  • End-to-end ML system design and deployment

Certifications

  • IBM – Apache Hadoop Developer
  • MathWorks – Machine Learning & MATLAB Onramps
  • AWS – Certified AI Practitioner (Upskilling)

Languages and Tools


GitHub Stats


Connect with Me

Pinned Loading

  1. Cognee-ML-GraphQA Cognee-ML-GraphQA Public

    Reflects your tool (Cognee), the domain (ML), and the implementation (Knowledge Graph).

    HTML 1

  2. Agentic-AI-with-AutoGen-Microsoft Agentic-AI-with-AutoGen-Microsoft Public

  3. harry-potter-qna-with-cognee harry-potter-qna-with-cognee Public

    Exploring How Cognee's Knowledge Graphs Can Answer Questions About the Harry Potter Universe

    Jupyter Notebook

  4. memvid-dgca-aircrash-ai memvid-dgca-aircrash-ai Public

    MemVID-DGCA-Aircrash-AI is an AI-powered knowledge extraction and Q&A system built on official aircraft accident investigation reports published by the Directorate General of Civil Aviation (DGCA),…

    Jupyter Notebook 1

  5. image-generation-models image-generation-models Public

    Trying on different Image Generation Models

    Python