class Pavan:
def __init__(self):
self.name = "Atchyuta Pavan Karthikeya"
self.role = "AI & ML Undergraduate"
self.degree = "B.Tech CSE (AI & ML), 2024–2028"
self.cgpa = 9.48
self.rank = "🥇 Rank 1 — CSE (AI & ML) Branch"
self.identity = "Hackathon Enthusiast 🏆"
self.focus = [
"Machine Learning 🧠",
"Generative AI ✨",
"Computer Vision 👁️",
"ML Ranking Systems 📊",
"Production-deployed AI 🚀"
]
self.currently_exploring = "Edge AI + Multi-Agent Systems"
self.fun_fact = "I turn hackathon ideas into deployed products"
self.philosophy = "Learn from every experience — always a continuous learner"
def say_hi(self):
return "Thanks for stopping by — let's build something! 🤝"
me = Pavan()
print(me.say_hi())- 🔭 Currently building AI-powered, production-deployed systems across GenAI, Computer Vision & ML ranking
- 🏆 Competing in national-level hackathons: Flipkart GRiD 8.0, Redrob AI National Hackathon, NHAI Hackathon 7.0
- 🥇 Top ~0.36% (95.57/100) — Hack2Skill × Google PromptWars Challenge 3, among 30,787+ participants
- 🥈 Selected for Round 2 — Meta AI / Hugging Face / PyTorch OpenEnv Hackathon among 31,000+ teams
- 💡 I learn fast, ship fast, and love turning a vague hackathon prompt into a real, working product
- 🌱 I believe every experience teaches something — I'm a continuous learner, always growing through what I build and what I get wrong
- 📫 Reach me at pavankarthikeyaatchyuta@gmail.com
| Project | What it does |
|---|---|
| 🌍 EcoVerse AI | Behavioral climate companion using Gemini — turns daily habits into a 50-year future-city simulation. Built for PromptWars Challenge 3. Live on Cloud Run. |
| 🧠 Redrob Candidate Ranker | Explainable 7-stage ML ranking pipeline — ranks 100K candidates in ~78s with honeypot/fraud detection, CPU-only. |
| 🗳️ Election Guide Assistant | Civic-AI assistant with a 7-stage state machine, Hindi TTS, and voice input — deployed on Google Cloud Run. |
| 🛡️ FaceGuard Offline | Fully offline facial recognition + liveness detection for field personnel, built with ONNX Runtime Mobile. |
| 📄 AI Resume Analyzer Pro | Recruiter-simulation system scoring resumes via structured Gemini prompts with JSON-mode output. |
| 📚 Study Bot | FastAPI + Gemini-powered learning assistant with MongoDB-backed chat history. |
- 🥇 Top 0.36% — Hack2Skill × Google PromptWars Challenge 3 (95.57/100, among 30,787+ participants) — EcoVerse AI
- 🥈 Round 2 Selection — Meta AI / Hugging Face / PyTorch OpenEnv Hackathon, among 31,000+ teams — Scaler School of Technology
- 🎓 Shortlisted — Amazon ML Summer School 2026
- 🏁 Participant — Flipkart GRiD 8.0 (AI Engineering Track) · Redrob AI National Hackathon · NHAI Hackathon 7.0
- 📜 Certified — Oracle OCI AI Foundations · NPTEL Python for Data Science & IoT · Google AI Essentials · Qualcomm AI Upskilling
