Skip to content
View mr-pravin's full-sized avatar
:electron:
𝐏𝐫𝐞𝐬𝐬𝐮𝐫𝐞 𝐈𝐬 𝐏𝐫𝐢𝐯𝐢𝐥𝐞𝐠𝐞
:electron:
𝐏𝐫𝐞𝐬𝐬𝐮𝐫𝐞 𝐈𝐬 𝐏𝐫𝐢𝐯𝐢𝐥𝐞𝐠𝐞

Block or report mr-pravin

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 supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mr-pravin/README.md
Typing SVG
INFINEURONITY Logo



Portfolio LinkedIn GitHub PyPI Play Store Email


coding gif

⚡ whoami

pravin = {
    "role"      : "AI/ML Engineer & Founder",
    "company"   : "INFINEURONITY  ∞  (Google Play)",
    "focus"     : [
        "GraphRAG · Knowledge Graphs",
        "Agentic LLM Systems",
        "Semantic Retrieval",
    ],
    "shipped"   : "1,000+ Android users · 2 PyPI libs",
    "stack"     : "Python · LangChain · FAISS · n8n · Gemini",
    "location"  : "Chennai, India 🇮🇳",
    "motto"     : "If it can't run at 50MB, engineer it until it can.",
}


GitHub Streak GitHub Stats
Top Languages by Repo Productive Time
Profile Details

⚡ Tech Arsenal

〔 Core Engineering 〕

Python Linux Git GitHub Flask FastAPI Docker MySQL VSCode Vercel Postman

〔 ML · GenAI · Agentic Systems 〕

LangChain FAISS GraphRAG HuggingFace Sentence Transformers n8n Gemini TensorFlow Scikit-learn OpenCV MediaPipe Pydantic NumPy Pandas Matplotlib Seaborn Jupyter Streamlit Prompt Engineering RAG Embeddings Quantization REST APIs Serial Communication Weights & Biases


∞ Flagship Builds

🧠 INFINEURONITY — Personal Memory AI

GraphRAG · FAISS · 4-bit Quantization · Android

Production AI memory engine live on Google Play. Processes Notes & PDFs via graph-structured RAG, retrieves context in 200ms within a tight 50MB on-device budget using sentence-transformers + quantized LLMs.

Users Latency Budget

Google Play

🔒 Guardian UPI — Fraud Detection AI

ML · Flask · Real-time API · 96% Precision

Deployed a production fraud detection API that processes live UPI transactions in real time. Built with scikit-learn + Flask, serving inference with 96% precision. Internship project turned live product.

Precision Status

Live Demo Source

🤖 AGIE — Agentic Growth Intelligence Engine

Gemini · n8n · Multi-step LLM Agents

Fully autonomous content pipeline: scrapes trends → LLM drafts SEO LinkedIn posts + YouTube scripts → auto-publishes. Zero human in the loop. Cuts content production by 80%.

Saved Stack

Source

✋ HumanIO — Gestured Bionic Hand

OpenCV · MediaPipe · Serial Communication

Real-time gesture recognition on 21-point hand landmarks achieving 94% accuracy at 30 FPS on CPU-only hardware — translating gestures into serial commands controlling a bionic hand prototype.

Accuracy FPS

Source

📦 mrpravin — One-Line AutoML (PyPI)

AutoML · Scikit-learn · Pydantic

pip install mrpravin — raw CSV in, deployment-ready model out. Auto-cleans, encodes, scales, cross-validates, and serializes — all in a single function call.

PyPI Version Downloads

📚 Think-Python-In-Tanglish — Python Book

Open-Source Education · Tamil + English

Authored a Python learning book in Tanglish (Tamil + English) to lower the barrier to programming for Tamil-speaking learners. 300+ copies distributed — democratizing CS education.

Copies Language

Read


📈 Contribution Activity

Activity Graph


🐍 Contribution Snake

GitHub contribution grid snake animation

📊 By The Numbers

╔═══════════════════════════════════════════════════════════════════════╗
║                                                                       ║
║   📱   1,000+    Android production users  (INFINEURONITY)            ║
║   📦       2     Published PyPI libraries                             ║
║   ⚡   200 ms    On-device RAG retrieval latency                      ║
║   💾    50 MB    Mobile memory budget for full ML pipeline            ║
║   🔒     96 %    UPI fraud detection precision  (live in prod)        ║
║   🎯     94 %    Real-time gesture recognition accuracy               ║
║   🏃    30 FPS   CV inference on CPU-only hardware                    ║
║   📝     80 %    Content creation time saved via AGIE                 ║
║   📚    300+     Tamil learners reached via Think-Python-In-Tanglish  ║
║   📈  1,047      GitHub contributions in 2025                         ║
║                                                                       ║
╚═══════════════════════════════════════════════════════════════════════╝

🏆 Certifications & Achievements

Credential Issuer Verify
Data Analytics Professional Meta · Coursera ✅ Certificate
Data Science, AI & Development IBM · Coursera ✅ Certificate
Claude Code in Action Anthropic · Skilljar ✅ Certificate
ML Internship DLK Software Solutions 🔗 Live Project
Python Developer Internship Ability Coding STU-241020-462987

Profile Views Followers


"Every millisecond shaved, every megabyte saved — that's the craft."

Popular repositories Loading

  1. Python Python Public

    commit

    Jupyter Notebook 2

  2. LyricsAnalyzer LyricsAnalyzer Public

    Generates original song lyrics using an LSTM neural network trained on a dataset of pop songs. Implements Flask API for web-based lyric generation.

    Python 2

  3. HumanIO HumanIO Public

    Real-time hand and eye gesture recognition system to control Arduino-driven LEDs using OpenCV, MediaPipe, and Python.

    Python 2

  4. Portfolio Portfolio Public

    TypeScript 2

  5. Think-Python-In-Tanglish Think-Python-In-Tanglish Public

    2

  6. UPI UPI Public

    Jupyter Notebook 2