class SiddharthPatel:
def __init__(self):
self.role = "AI Engineer & ML Researcher"
self.education = "B.Tech CS (AI/ML) | Bennett University"
self.cgpa = "9.66/10 | Dean's List (3x)"
self.location = "Greater Noida, India"
def current_work(self):
return {
"company": "Cograd Technologies",
"position": "AI Intern",
"focus": ["Multi-Agent RAG Systems", "LLM Pipelines",
"Vector Databases", "Azure Cloud Deployment"],
"impact": "Serving 10K+ users"
}
def research_interests(self):
return [
"🧠 Large Language Models (LLMs)",
"🔬 Physics-Informed Neural Networks (PINNs)",
"👁️ Computer Vision & Medical Imaging",
"🤖 Multi-Agent AI Systems",
"🛡️ AI Safety & Constitutional AI"
]
def achievements(self):
return {
"publications": 3,
"hackathons": "Top 1% Amazon ML | Top 50 Pan-IIT",
"gate_2025": "Top 10% (Data Science & AI)",
"projects_deployed": "Production-grade RAG systems"
}
me = SiddharthPatel()
print(f"Currently: {me.current_work()}")
🎯 AI-Powered Live Class Doubt Management • Dec 2025 - Present
- Real-time NLP system with 70% faster response using pgvector clustering
- Async LLM pipeline with 85%+ accuracy for context-aware answers
- Handles 100+ concurrent doubts via Chrome extension + FastAPI
🔮 DataWhiz • Sep 2025 - Jan 2026 • Live Demo
- Novel multi-agent Text-to-SQL system for 200+ table databases
- Multi-agent orchestration with auto-correction deployed on Azure
- Automated visualization using LIDA for enterprise analytics
💎 Aurigen • Feb 2025 - Apr 2025
- SDXL + ControlNet pipeline with custom LoRA fine-tuning (6K images)
- 3x inference speedup via FP16 quantization & attention caching
[AICAPS 2026] Hinglish Abusive Comment Detection Using Transformer-Based Models
→ Hybrid mBERT/XLM-R + BiGRU • 90% accuracy on 700K+ code-mixed posts
[IC3SE 2025] Deep Learning-based Brain Tumor Detection
→ Multimodal MRI CNN • 94% accuracy with improved interpretability
[MAC 2024 - IEEE] CNN-Based Skin Disease & Cancer Classification
→ 96.64% accuracy across 57 disease classes • Read Paper
🥇 Top 1% Amazon ML Challenge 2024 (30K+ teams)
🥈 Top 50 Convolve 3.0 - Pan IIT AI/ML Hackathon (4K+ teams)
🥇 1st Place Inspire Awards (6K+ teams)
🎓 Dean's List (3 semesters) | Top 10% GATE 2025
📜 Amazon ML Summer School 2025 (Selected nationwide)
areas:
- AI Safety & Constitutional AI
- Reinforcement Learning from Human Feedback (RLHF)
- Multi-Modal Large Language Models
- Edge AI & Model Optimization
- Graph Neural Networks (GNNs)
