┌──────────────────────────────────────────────────────────┐
│ nomen : Akash Nagar │
│ munus : AI/ML Engineer & Agentic AI Architect │
│ locus : India 🇮🇳 │
│ status : Fabricating Real-World AI Systems 🚀 │
│ │
│ specializatio: │
│ ├── Machine Learning ✅ Consummated │
│ ├── Agentic AI (n8n) ✅ Preternatural │
│ ├── LangGraph ✅ Proficient │
│ ├── AWS + Data Pipelines ✅ Empirically Hands-On │
│ ├── Deep Learning ⏳ 20% Inchoate │
│ ├── Generative AI / GANs ⏳ 25% Inchoate │
│ └── NLP & Transformers ⏳ Inchoate │
│ │
│ passion : Football ⚽ | Peregrination 🌍 | Cars 🚗 | Anime 🎌 │
└──────────────────────────────────────────────────────────┘I am an AI/ML engineer hailing from India who fabricates production-ready AI systems that actually eventuate. I have consummated Machine Learning end-to-end and operate at a preternatural echelon with Agentic AI — automating labyrinthine workflows via n8n, LangGraph, and LLM APIs. I have constructed pipelines on AWS (S3 → Snowflake + data visualization), and I am indefatigably advancing in Deep Learning and Generative AI.
These are my most labyrinthine, end-to-end production systems — each one evinces a distinct facet of my engineering profundity.
|
The full stack, zero-touch analytics pipeline. Percuss a plain English interrogation → system auto-generates SQL → executes it → feeds results to an LLM to compose a human-readable business report → auto-delivers the report directly to WhatsApp. No dashboards. No analysts. Just resolutions. Architecture Flow: Impact: Replaces an entire analytics workflow. Any non-technical user can interrogate a database via WhatsApp in natural language. |
|
|
Full-stack e-commerce platform with built-in AI analytics. A production-grade e-commerce web application with a dedicated analytics tab powered by LLM insights. Built with Flask + Python backend, MySQL Connector for data persistence, and an integrated LLM API that auto-interprets sales trends, customer behavior, and inventory patterns. Key Features: Stack: Python · Flask · MySQL Connector · LLM API · Chart.js · HTML/CSS/JS |
|
Enterprise-grade cloud data engineering pipeline, end-to-end. A production data pipeline constructed entirely on AWS: raw dataset uploaded to S3, transformed and loaded into Snowflake via AWS Glue/Lambda, then visualized live in a Streamlit dashboard connected directly to Snowflake — all streaming in real time. Pipeline Architecture: Highlights: Demonstrates full cloud data engineering stack. From raw CSV to interactive dashboard — zero local infrastructure required. |
|
|
Real-world Deep Learning + full MLOps lifecycle. Deployed. An end-to-end disease detection system that classifies chicken fecal images as Coccidiosis or Healthy using a CNN-based deep learning model. This project covers the complete ML lifecycle — from raw data collection all the way to containerized deployment. Full MLOps Pipeline: Why it matters: Applies Deep Learning to solve a real agricultural problem. Full MLOps — not just a Jupyter notebook. |
$ ./show_progress.sh --full
╔══════════════════════════════════════════════════════════════╗
║ ERUDITIONIS PROGRESSUS CHART ║
╠══════════════════════════════════════════════════════════════╣
║ ║
║ CONSUMMATED ✅ ║
║ ───────────────────────────────────────────────────────── ║
║ [██████████] Machine Learning 100% CONSUMMATED ║
║ [██████████] Feature Engineering 100% CONSUMMATED ║
║ [██████████] Data Cleaning & EDA 100% CONSUMMATED ║
║ [██████████] Regression & Classification 100% CONSUMMATED ║
║ [██████████] Ensemble Learning 100% CONSUMMATED ║
║ [██████████] SQL & Database Operations 100% CONSUMMATED ║
║ [██████████] Python for Data Science 100% CONSUMMATED ║
║ [██████████] Agentic AI with n8n 100% PRETERNATURAL ⭐║
║ [██████████] LangGraph 100% PROFICIENT ⭐ ║
║ [██████████] AWS (S3, Snowflake, Viz) 100% HANDS-ON ⭐ ║
║ ║
║ INCHOATE ⏳ ║
║ ───────────────────────────────────────────────────────── ║
║ [██░░░░░░░░] Deep Learning ~20% INCHOATE ║
║ [███░░░░░░░] Generative AI / GANs ~25% INCHOATE ║
║ [██░░░░░░░░] NLP & Transformers ~20% INCHOATE ║
║ [██░░░░░░░░] Computer Vision ~15% INCHOATE ║
║ ║
║ IN EXSPECTO 🔲 ║
║ ───────────────────────────────────────────────────────── ║
║ [░░░░░░░░░░] MLOps & CI/CD 0% IN EXSPECTO ║
║ [░░░░░░░░░░] Apache Spark / Kafka 0% IN EXSPECTO ║
║ [░░░░░░░░░░] Databricks / Snowflake 0% IN EXSPECTO ║
║ ║
║ UNIVERSALIS ████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ ~45% ║
╚══════════════════════════════════════════════════════════════╝
Real systems. Real automation. Real impact.
| 🏆 | Opus | Descriptio | Tech Stack | Status |
|---|---|---|---|---|
| 🔴 | Anglicē-ad-SQL + LLM Relatio + WhatsApp Traditio | Natural language → auto SQL → LLM report → WhatsApp delivery. Full zero-touch analytics pipeline | n8n, LLM, SQL, WhatsApp API | ✅ Consummated |
| 🔴 | E-Commerce Web App + Analytics Dashboard | Full e-commerce site with LLM-powered analytics tab, real-time KPIs, and MySQL backend | Flask, Python, MySQL, LLM API | ✅ Consummated |
| 🔴 | AWS S3 → Snowflake → Streamlit Pipeline | End-to-end cloud data pipeline: S3 ingestion → Snowflake warehouse → live Streamlit dashboard | AWS S3, Snowflake, Streamlit | ✅ Consummated |
| 🔴 | Chicken Coccidiosis Detector — DL + MLOps | CNN-based fecal image classifier (Coccidiosis/Healthy) with full MLOps: DVC, MLflow, Docker, CI/CD | PyTorch, MLflow, DVC, Docker | ✅ Consummated |
| ⭐ | AI + Humanizer LinkedIn Post Automator | AI writes posts, humanizes content, attaches generated images, auto-publishes to LinkedIn | n8n, LLM, Image Gen, LinkedIn API | ✅ Consummated |
| ⭐ | WhatsApp Agent (Grok-3 / xAI) | Conversational WhatsApp agent powered by Grok-3, built end-to-end in VS Code | Python, Grok API, WhatsApp | ✅ Consummated |
| ⭐ | AI Appointment Agent (WhatsApp) | Fully autonomous appointment booking agent over WhatsApp | n8n, LLM, WhatsApp | ✅ Consummated |
| 🔥 | Review Analysis System | Automated review scraping, sentiment analysis & reporting pipeline | n8n, LLM, Python | ✅ Consummated |
| 🔥 | AI Blog Generator → Email Delivery | AI writes SEO blog posts and auto-sends to mailing list | n8n, LLM, Gmail API | ✅ Consummated |
| 🔥 | Lead Generation Engine | Generated 50+ warm leads via smart WhatsApp outreach automation | n8n, WhatsApp API | ✅ Consummated |
| 🔥 | WhatsApp Messaging Web App | Web app with contact form, instant WhatsApp delivery via Twilio + full message history dashboard | Python, Flask, Twilio | ✅ Consummated |
| 💡 | Python Interview Bot (Telegram) | Automated Python interview Q&A bot with Groq API | n8n, Groq API, Telegram | ✅ Consummated |
| 💡 | WhatsApp Pharmacy Manager | Full pharmacy management automation system | n8n, WhatsApp API | ✅ Consummated |
| 💡 | Gmail Auto Reply System | Intelligent automated email response workflow | n8n, Gmail API | ✅ Consummated |
| 💡 | GitHub Daily Notes Automator | n8n workflow that commits random notes.md files to GitHub daily | n8n, GitHub API | ✅ Consummated |
| 📊 | ML Models Collection | Multiple ML algorithms & end-to-end implementations | Python, sklearn | ✅ Consummated |
| 🚀 | Data Science Notes Repo | Complete ML & DS structured learning notes | Jupyter, Python | 🚀 Active |
| 🚀 | Deep Learning Repository | DL, NLP & CV implementations in progress | PyTorch, TF | 🚀 Active |
🔴 = Flagship Priority Projects | ⭐ = High Impact | 🔥 = Production Agentic Systems
🔥 Profunda Eruditio
- ANN · CNN · RNN · LSTM · Transformers
- Backpropagation · Optimization Techniques
- Batch Norm · Dropout · Regularization
👁 Visio Computatralis
- OpenCV · YOLO · Object Detection
- Image Segmentation · Sports AI Projects
- Real-Time CV Applications
💬 NLP & LLMs
- BERT · GPT · Attention Mechanisms
- RAG Systems · AI Chatbots · AI Assistants
- Fine-tuning & Prompt Engineering
🤖 Agentic AI — ✅ Preternatural
- n8n · LangChain · LangGraph ← Already Proficient
- Multi-Agent Systems · Tool Use · Memory
- Complex workflow automation with LLMs & APIs
- Next: advanced multi-agent orchestration patterns
☁️ Cloud & Deployment — ✅ Hands-On
- AWS S3 → Snowflake pipelines ← Built & Deployed
- Data Visualization on AWS ← Done
- Streamlit Dashboards ← Done
- Flask · Docker
- Next: MLOps, CI/CD, advanced AWS services
📦 Big Data & Platforms (In Exspecto)
- Hadoop · MapReduce · Apache Hive · Apache Spark
- Databricks · Snowflake
- ETL Concepts · Apache Kafka
- Data Lakes · Data Warehousing
- Power BI · AWS
| 🎯 Opus | 🛠️ Technologiae | Status |
|---|---|---|
| Handwritten Word Recognition | CRNN (CNN + LSTM + CTC) | 🔲 In Exspecto |
| Sports AI with Real-Time Tracking | YOLO, OpenCV, CV | 🔲 In Exspecto |
| Hand Gesture Recognition App | Computer Vision, Deep Learning | 🔲 In Exspecto |
| GAN Image Generator | GANs, PyTorch | 🔲 In Exspecto |
| Multi-Agent Research Assistant | LangGraph, RAG, LLMs | 🔲 In Exspecto |
→ Become a Professional AI/ML Engineer
→ Build Advanced Multi-Agent AI Systems
→ Create Real-World Generative AI Products
→ Deploy End-to-End MLOps Pipelines
→ Launch an AI Startup Project
→ Secure Remote AI/ML Opportunities Globally
⚙️ Click to setup Snake Animation (GitHub Actions Workflow)
Create .github/workflows/snake.yml in your techakash32/techakash32 repository:
name: Generate Snake
on:
schedule:
- cron: "0 0 * * *"
workflow_dispatch:
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Generate Snake
uses: Platane/snk@v3
with:
github_user_name: techakash32
outputs: |
dist/github-snake.svg
dist/github-snake-dark.svg?palette=github-dark
- name: Push to output branch
uses: crazy-max/ghaction-github-pages@v3.1.0
with:
target_branch: output
build_dir: dist
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}After adding, go to Actions tab → run the workflow manually once.
⚽ Football | 🌍 Peregrination | 🚗 Cars | 🎌 Anime
|
 |
 |
 |