══════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════
██╗ ██╗████████╗███████╗ █████╗ ██╗ ██╗
██║ ██║╚══██╔══╝██╔════╝██╔══██╗██║ ██║
██║ ██║ ██║ ███████╗███████║██║ ██║
██║ ██║ ██║ ╚════██║██╔══██║╚██╗ ██╔╝
╚██████╔╝ ██║ ███████║██║ ██║ ╚████╔╝
╚═════╝ ╚═╝ ╚══════╝╚═╝ ╚═╝ ╚═══╝
███╗ ███╗███████╗██╗ ██╗████████╗ █████╗
████╗ ████║██╔════╝██║ ██║╚══██╔══╝██╔══██╗
██╔████╔██║█████╗ ███████║ ██║ ███████║
██║╚██╔╝██║██╔══╝ ██╔══██║ ██║ ██╔══██║
██║ ╚═╝ ██║███████╗██║ ██║ ██║ ██║ ██║
╚═╝ ╚═╝╚══════╝╚═╝ ╚═╝ ╚═╝ ╚═╝ ╚═╝
DATA ENGINEER • ML ENGINEER • CLOUD ARCHITECT
Building Systems That Scale to Millions
══════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════
I don't build to learn. I learn to build.
Every line of code ships to production. Every project solves real problems. Every metric tells a story.
What I Actually Do:
- 🏗️ Architect data pipelines processing millions of records with zero downtime
- 🤖 Deploy ML models that drive real revenue and reduce operational costs
- ☁️ Build zero-cost cloud infrastructure because smart > expensive
- 📊 Create real-time analytics that executives actually use
- 🛡️ Automate security compliance so engineers can focus on shipping
Tech I Ship With:
Python PySpark Kafka PostgreSQL AWS Docker Ansible React Flask Node.js
Location: India 🇮🇳 | Status: Building in public | Response time: <24 hours
class UtsavMehta:
def __init__(self):
self.role = "Data Engineer & ML Engineer"
self.location = "India 🇮🇳"
self.status = "Building in public"
self.motto = "Ship fast, scale smart, measure everything"
def what_drives_me(self):
return {
"impact": "70% faster processing, 85% less manual work",
"efficiency": "$0 cloud costs, infinite scalability",
"reality": "Production systems, not portfolio projects"
}
def current_mission(self):
return [
"Real-time streaming with Kafka (because batch is dead)",
"MLOps that actually works (not just notebooks)",
"Distributed systems at scale (millions of events/sec)"
]|
The Problem: Retail chaos — POS systems, e-commerce orders, inventory scattered across 5 different databases. The Solution: Full-scale ETL pipeline with Pydantic validation, PySpark transformation, Star Schema warehouse. The Numbers: What Makes It Different:
|
The Problem: Manual security audits taking 8 hours/week. Compliance chaos. The Solution: Automated compliance engine with real-time monitoring, live dashboards, infrastructure as code. The Numbers: What Makes It Different:
|
|
The Problem: Education lenders losing money on high-risk loans. No data-driven decisions. The Solution: ML engine predicting employment outcomes by analyzing student profiles + placement history + job market signals. The Numbers: What Makes It Different:
|
The Problem: 10,000+ customer records. Zero insights. Decisions based on "gut feeling." The Solution: Statistical analysis pipeline with hypothesis testing, correlation analysis, business recommendations. The Numbers: What Makes It Different:
|
📊 Data & Analytics Projects (4 projects)
| Project | What It Does | Impact | Tech Stack |
|---|---|---|---|
| Student Social Media Analytics | Power BI dashboard mapping social media addiction vs academic performance | Identified correlation between screen time & grades | Power BI, DAX, Data Modeling |
| SalesDataAnalysis | Amazon sales dashboard with VBA automation | Automated weekly reporting (saved 5 hrs/week) | Excel, VBA, Pivot Tables |
| FundFlow | Financial analysis & allocation platform | Portfolio optimization algorithms | Python, Jupyter, Pandas |
| Retail Store SQL | Progressive SQL learning (6 difficulty levels) | Educational resource for SQL learners | PostgreSQL, Complex Queries |
🤖 AI & Machine Learning Projects (3 projects)
| Project | What It Does | Impact | Tech Stack |
|---|---|---|---|
| Tiny-LLM | Lightweight LLM exploring modern AI architecture | Understanding transformer internals | JavaScript, NLP, Attention |
| final-project-emb-ai | Embedded AI for web applications | Real-time inference in browser | HTML, JavaScript, TensorFlow.js |
| Data-Encryption | Cryptography web app with QR encoding | ⭐ 1 Star | Flask, Cryptography, QR |
💻 Full Stack Applications (4 projects)
| Project | What It Does | Impact | Tech Stack |
|---|---|---|---|
| FinTrack | Personal finance tracker with expense management | Budget planning & insights | JavaScript, Node.js, MongoDB |
| E-Api | Enterprise-level backend API service | RESTful architecture | Node.js, Express, JWT |
| First-Website | Journey into web development | Portfolio piece | HTML, CSS, JavaScript |
| 72232885M | Advanced JavaScript algorithms | Problem-solving showcase | JavaScript, Algorithms |
Languages I Think In:
Python JavaScript Java Kotlin SQL Bash
Data Engineering:
PySpark Apache Kafka PostgreSQL MongoDB Redis Airflow Pandas NumPy
Cloud & Infrastructure:
AWS Docker Kubernetes Ansible Terraform Nginx LocalStack CI/CD
ML & Analytics:
Scikit-learn TensorFlow PyTorch MLflow Power BI Jupyter Statistical Modeling
Web & APIs:
React Flask Node.js FastAPI REST GraphQL WebSockets
📊 16+ Production Projects | 🔥 4 Featured Systems | ⚡ 5+ Tech Stacks Mastered
💰 $0 Cloud Costs | 📈 70% Faster ETL | 🛡️ 85% Less Manual Work
| Traditional Approach | My Approach |
|---|---|
| "I'm learning React" | "I shipped a React app serving 10K users" |
| "Here's my tutorial project" | "Here's my production system with metrics" |
| "AWS is expensive" | "I built zero-cost cloud with LocalStack" |
| "I know Python" | "I built a PySpark pipeline processing 5M records/day" |
| "I'm interested in ML" | "I deployed ML that increased accuracy by 40%" |
+ Ship to production, not to portfolio
+ Measure everything (if you can't measure it, it didn't happen)
+ Learn by building (not by watching tutorials)
+ Real business problems > Toy projects
+ Business impact > Technical complexity for complexity's sakeExecution Speed: I ship MVPs in weeks, not months. When everyone's still planning, I'm already iterating based on real data.
Zero-Cost Mindset: Built production infrastructure with $0 cloud costs using LocalStack, Docker, and smart architecture. Resourcefulness > budget.
Business First: Every technical decision traces back to a business metric. I don't build for the sake of building—I build to move needles.
Production Obsessed: 16+ projects in production. Not in "development." Not "almost done." Shipped. Used. Measured.
🔨 Building:
- Real-time streaming pipeline with Apache Kafka (handling 1M events/sec)
- MLOps platform with automated retraining & monitoring
- Distributed tracing system for microservices
📚 Learning:
- Advanced distributed systems patterns
- Rust for high-performance data processing
- Kubernetes + service mesh architecture
✍️ Writing on Medium:
- "How I Built Zero-Cost Cloud Infrastructure"
- "ML in Production: What They Don't Teach You"
- "Data Engineering: From Local to Millions"
🎯 2026 Goals:
[✓] Ship 16+ production projects
[▓▓▓▓▓▓▓▓▓░] Contribute to 5 open-source projects (4/5)
[▓▓▓▓▓▓░░░░] Publish 12 technical articles (7/12)
[▓▓▓░░░░░░░] Build production ML system (30%)
[░░░░░░░░░░] Master Rust (starting)
[░░░░░░░░░░] AWS Solutions Architect cert (Q4)
I'm not looking for "opportunities." I'm looking for ambitious projects that actually matter.
For Companies:
- Data Engineering roles building systems that scale
- ML Engineering positions shipping production AI
- Cloud Architecture projects (AWS/K8s/distributed systems)
- Technical leadership roles where I can build teams & culture
For Founders:
- Technical co-founder for data-intensive startups
- Advisory roles for ML/data strategy
- Contract work building MVP data infrastructure
For Community:
- Open-source contributions to data engineering tools
- Technical writing & knowledge sharing
- Mentoring engineers who want to ship real products
Systems that process millions of events. ML that drives real revenue. Infrastructure that costs $0 but scales infinitely. Projects where metrics matter and impact is measurable.
Systems processing millions of events daily. ML models driving measurable revenue. Infrastructure that costs $0 but scales infinitely. Projects where every line of code ships and every metric matters.
Best way to reach me: Twitter DM or LinkedIn message. I respond within 24 hours.
┌─────────────────────────────────────────────────────────────────────┐
│ │
│ "I don't build to learn. I learn to build." │
│ │
│ Every project ships to production. │
│ Every metric tells a story. │
│ Every line of code solves a real problem. │
│ │
│ Let's build something that matters. │
│ │
└─────────────────────────────────────────────────────────────────────┘
⚡ Last Updated: July 2026 | 🚀 Built With: Python, Coffee & Obsession | 💙 Made By: Utsav Mehta