class AyushDubey:
def __init__(self):
self.role = "AI/ML Engineer"
self.location = "Abu Dhabi, UAE 🇦🇪"
self.expertise = [
"Generative AI (LLMs, RAG, Agents)",
"Deep Learning & Computer Vision",
"MLOps & Production Deployment",
"Full-Stack AI Applications"
]
self.current_focus = "Building scalable agentic AI systems"
self.achievement = "30% cost reduction via optimized ML pipelines"
def say_hi(self):
print("Thanks for dropping by! Let's build something amazing together!")
me = AyushDubey()
me.say_hi()🔥 What Sets Me Apart
- 🎯 Production-First Mindset: 10+ deployed models in real-world applications
- 💡 Cost Optimizer: Achieved 30% cost reductions through intelligent architecture
- 🤖 Agentic AI Expert: Built multi-agent systems with CrewAI, LangChain & LlamaIndex
- 🚀 Full-Stack Capability: From model training to containerized deployment
- 📊 Proven Track Record: B.Tech AIML (8.20 GPA) + Multiple Production Systems
|
Fullstack AI Engineer • Sept-Oct 2025
|
AI Consultant • Oct 2024-Jan 2025
|
|
AI Developer • Dec 2023-June 2024
|
Data Scientist • June 2023-Feb 2024
|
🎯 Production-grade RAG system with multi-agent architecture
├─ ⚡ Sub-second retrieval using LlamaIndex + ChromaDB
├─ 🐳 Containerized microservices (Docker + FastAPI + Streamlit)
├─ 🤖 Multi-agent orchestration with CrewAI
├─ 🦙 Groq Llama 3.1 inference for scalable deployment
└─ ✅ Pydantic validation for conversational context management
Tech Stack: LlamaIndex • ChromaDB • FastAPI • Streamlit • CrewAI • Docker • Groq
🎯 Multi-agent AI system for automated equity research
├─ 👔 Analyst Agent: Analyzes 15+ financial metrics
├─ 📊 Trader Agent: Generates actionable trading strategies
├─ 📉 Real-time data integration (Yahoo Finance API)
├─ 🔄 90% reduction in manual analysis time
└─ 🎲 100% reproducible trading recommendations
Tech Stack: Multi-Agent AI • Yahoo Finance API • LLMs • Custom Financial Tools
🎯 CNN system beating DeepFood benchmark on Food101 dataset
├─ 🎯 85% accuracy (vs DeepFood's 77.4%)
├─ ⚡ 90-minute training (vs 3 days) using Mixed Precision
├─ 🧠 EfficientNetB1 with 40% memory reduction
├─ 🚀 Real-time Streamlit deployment with Top-5 predictions
└─ 📊 Optimized pipeline with adaptive learning rates
Tech Stack: TensorFlow/Keras • EfficientNetB1 • Streamlit • Mixed Precision Training
|
B.Tech in AI & Machine Learning |
|
%%{init: {'theme':'dark'}}%%
graph LR
A[AI/ML Engineering] --> B[Generative AI]
A --> C[Deep Learning]
A --> D[MLOps]
B --> B1[LLMs & RAG]
B --> B2[Multi-Agent Systems]
B --> B3[Prompt Engineering]
C --> C1[Computer Vision]
C --> C2[NLP]
C --> C3[Time Series]
D --> D1[Docker/Kubernetes]
D --> D2[CI/CD Pipelines]
D --> D3[Monitoring Tools]
style A fill:#00D9FF,stroke:#00D9FF,color:#000
style B fill:#FF6B6B,stroke:#FF6B6B,color:#fff
style C fill:#4ECDC4,stroke:#4ECDC4,color:#000
style D fill:#FFD93D,stroke:#FFD93D,color:#000
| 🎯 Area | 📊 Status | 🔥 Intensity |
|---|---|---|
| Agentic AI Systems | Building production systems | 🟢🟢🟢🟢🟢 |
| RAG Optimization | Advanced retrieval techniques | 🟢🟢🟢🟢⚪ |
| MLOps at Scale | Kubernetes orchestration | 🟢🟢🟢🟢⚪ |
| LLM Fine-tuning | Domain-specific models | 🟢🟢🟢⚪⚪ |
I'm passionate about building production-grade AI systems that solve real-world problems.
Currently seeking opportunities to work on cutting-edge AI/ML projects.
✨ Full-time AI/ML Engineering roles
🚀 Generative AI & Agentic Systems projects
💼 Consulting on RAG pipelines & LLM deployment
🤖 Open-source collaborations
