class SakshiBhatia:
    def __init__(self):
        self.role        = "Software Engineer — Full-Stack & ML"
        self.education   = [
            "IIITDM Jabalpur — B.Tech Electronics & Communication Eng. (CPI 8.1)",
            "IIT Madras     — BS Data Science & Applications (Foundation)"
        ]
        self.stack       = ["React", "Next.js", "TypeScript", "FastAPI", "PostgreSQL", "Docker"]
        self.data_tools  = ["Power BI", "DAX", "Pandas", "NumPy", "Scikit-learn"]
        self.currently   = "Building ConfirmHoga — an AI-powered railway intelligence platform"
        self.also_into   = ["Competitive Programming", "Open Source", "Drone Systems", "Chess ♟"]
        self.open_to     = ["SWE Internships", "Open Source Collabs", "Research", "Hackathon Teams"]
        self.fun_fact    = "I debug backend race conditions and opening theory with equal intensity"

    def greet(self):
        print("Thanks for stopping by — let's build something that matters!")

me = SakshiBhatia()
me.greet()

🎯 I bridge hardware intuition (ECE) with software craftsmanship (Data Science + full-stack engineering) — bringing that cross-domain thinking into every project, from ML pipelines to production APIs.

📚 Coursework: Data Structures & Algorithms · Database Management Systems · Operating Systems · Computer Networks · System Design · OOP · Statistics

200+ problems solved across dynamic programming, graphs, and binary search. (Optional: add live LeetCode/Codeforces stat cards here once you confirm your handles — placeholders below.)

React TypeScript FastAPI Scikit-learn PostgreSQL (Supabase)

Predicts PNR confirmation probability and surfaces cost-saving "Jugaad" split-ticket routes for Indian Railways passengers.

• 🎯 Random Forest Regressor on historical booking data — 91.8% accuracy (R² 0.92, RMSE 0.11)
• ⚡ Failover/caching architecture (Tenacity, Cachetools) across multiple live rail data providers for reliability
• 🧠 Custom BFS graph-traversal to identify split-ticket combinations that save passengers money
• 🗄️ Real-time train/fare data served via structured SQL queries against Supabase/PostgreSQL

FastAPI Next.js Scikit-learn

Real-time AQI + water-quality intelligence across 50+ Indian cities, built at an MLH-affiliated national hackathon.

• ⚡ Sub-200ms response times at scale
• 🤖 4-tier LLM fallback chain (Gemini, Groq, DeepSeek) — 99% uptime through rate-limit spikes
• 📄 Scikit-learn pipeline extracting structured data from PDF reports — 87% precision on water-drinkability prediction

JavaScript Chrome APIs Gemini API

Kills context-switching for knowledge workers with AI-assisted task automation.

• 🧠 Gemini API integration for task automation, synced with Google Calendar
• 📈 150+ active users, latency reduced by 35% via async API orchestration

Python OpenCV Machine Learning

Real-time CV safety system tracking Eye Aspect Ratio (EAR) at 25+ FPS.

• 🔔 Automated alerts cut simulated reaction delay by 40%
• 🧪 Validated across 500+ simulated frames

🏛️ Student Senate Member Spearheaded student-led sustainability initiatives — reduced the college's environmental footprint by 40%.

📢 Abhikalpan '25 & Collage Fests — PR & Sponsorship Lead (IIITDM Jabalpur Annual Techfest) Contacted 50+ potential sponsors, driving a 35% increase in event funding.

🌱 Open Source Contributor — GirlScript Summer of Code (GSSoC) Resolved bugs and enhanced modular components across open-source repos; authored unit tests, delivered via peer-reviewed PRs.

💬 Open to SWE internships, open-source collaborations, and hackathon teams. If you're building something interesting, reach out.

"Build with intention. Ship with pride. Learn always."

_{ECE × Data Science × Open Source × Robotics × Chess ♟ · Made with 💙 by Sakshi}