PranavMishra17
PRANAV MISHRA
"In order to keep up with everything AI, you need to be unemployed.
In order to keep paying for the tokens, you need to be employed."
</div>
🚀 Currently Building
🎮 SoulEngine Stateless NPC intelligence with memory & MCP-based agency |
🎤 MockFlow-AI Real-time voice interview platform with sub-400ms latency |
🏆 Major Wins
🕴🏼 IEEE 2026 - (ACCEPTED) Research in enterprise RAG systems with Metadata Enrichment. Presented MetaRAG paper at Informs+ Conference at Indianapolis, IN, USA - 2025
| 🥇 SnAIder-Cut - Meta Quest 3 app | Won a special track at MIT XR Hackathon, Jan 2024, Boston, MA |
📝 ICML 2026 (under review) - Multi-agent system using SLMs with modular teamwork components Link ↗
📝 PAKDD 2026 (under review) - Big5 Agents framework submission as TeamMedAgents Link ↗
🥇 HINT 5.0 Hackathon Winner - Virtual van Gogh: Virtual art museum for NFT collections
✅ Built & Shipped
-
📰 Resume-Craft-Pro 🌟: A full-stack AI web app for Resume customization with intelligent agentic architecture Link ↗
-
🐍 SnakeAI-MLOps: Complete RL pipeline with 4 algorithms + deployment Link ↗
-
🗣️ Voice-Persona: Dataset for character consistency in voice synthesis, generated with audio-language model Qwen2-Audio-7B
-
🎨 Auto-PaintSeg: Diffusion models for automated image segmentation Link ↗
-
🌌 Stellarium: VR space odyssey with 107K+ real astronomical objects Link ↗
Resume + Portfolio
📄 Resume View / Download my latest resume PDF |
🌐 Live Portfolio Explore projects, research, and engineering work |
⚡ The Arsenal
class PranavMishra:
def __init__(self):
# Core Languages
self.languages = ["Python", "C++", "JavaScript/TypeScript", "C#", "Java", "Rust"]
# AI/ML Stack
self.ai_frameworks = ["PyTorch", "TensorFlow", "LangChain", "LangGraph", "LibTorch", "Transformers"]
self.ai_platforms = ["OpenAI API", "Anthropic Claude", "Google Gemini", "Azure OpenAI", "Hugging Face"]
self.mlops = ["vLLM", "Docker", "GitHub Actions", "Model Versioning", "Experiment Tracking"]
self.specialties = [
"Multi-Agent Systems",
"RAG Architectures",
"Computer Vision",
"Reinforcement Learning",
"NLP/LLMs",
"Small Language Models",
"Healthcare AI"
]
# Full-Stack & Infrastructure
self.web_stack = {
"frontend": ["React", "Next.js", "TypeScript", "Tailwind"],
"backend": ["FastAPI", "Node.js", "Flask", "Django", "Express.js"],
"realtime": ["WebSockets", "Azure Speech Services", "Streaming APIs"]
}
self.databases = ["PostgreSQL", "MongoDB", "Pinecone", "CosmosDB", "Redis", "pgvector"]
self.cloud_platforms = ["Azure", "AWS (EC2, S3, Lambda)", "GCP", "Railway", "Vercel"]
self.devops = ["Docker", "Kubernetes", "CI/CD Pipelines", "Git", "Infrastructure as Code"]
# Game/XR Development
self.game_engines = ["Unity (C#)", "Unreal Engine 5 (C++)", "SFML", "WebXR"]
self.xr_experience = "MIT Reality Hack 2024 Winner 🏆"
# Research & Production Experience
self.current_work = {
"research": "UIC V-ARE Labs - Healthcare AI, VR/AR Medical Training",
"industry": "WheelPrice (Techstars) - Production CV/ML Systems",
"academic": "AAAI 2025 Submission (First Author) - Multi-Agent Medical AI"
}
# Notable Projects
self.portfolio_highlights = [
"TeamMedAgents: Multi-agent medical AI (7/8 benchmark improvements)",
"SLM-TeamMedAgents: Extended to Small Language Models (Google ADK)",
"MetaRAG: Enterprise RAG (10K+ docs/hour, Kafka pipelines)",
"IVORY: Production RAG platform on Azure (medical education)",
"ClauseCraft: Agentic document editor (Gemini + MCP)",
"Flow Planner: Autonomous UI workflow capture (Browser-Use + Claude)",
"SnakeAI-MLOps: Complete MLOps pipeline (5x training speedup)",
"Patient Care Automation: HIPAA-compliant healthcare workflows"
]