I'm a Machine Learning Engineer who loves turning complex AI concepts into production-ready applications. Whether it's training deep learning models, building multi-agent AI systems, or deploying full-stack ML apps — I thrive at the intersection of research and engineering.
class Arsh:
def __init__(self):
self.role = "ML Engineer & AI Developer"
self.focus = ["End-to-end ML Pipelines", "LLM Applications", "Multi-Agent Systems"]
self.currently_learning = ["Advanced RAG", "Agent Orchestration", "MLOps at Scale"]
self.fun_fact = "I believe the best code is code that ships 🚢"
def daily_routine(self):
return ["☕ Coffee", "🧠 Build AI", "🐛 Debug", "🚀 Deploy", "🔁 Repeat"]
|
|
|
|
Automatic CI/CD failure detection & AI-powered root-cause analysis
🔥 Key Features:
- Monitors GitHub Actions in real-time
- AI-generated failure analysis with Groq LLM
- Auto-creates GitHub issues with fix suggestions
- One-command setup:
autoops run
Tech: Python, GitHub API, Groq AI, CLI Tool
Multi-agent AI system with CEO & specialized teams
🔥 Key Features:
- CEO agent coordinating research & writing teams
- LangGraph workflow orchestration
- Modular agent architecture
- Complex task decomposition & execution
Tech: LangChain, LangGraph, Python
Transformer-based summarization with full MLOps pipeline
🔥 Key Features:
- T5 Transformer model for state-of-the-art summaries
- Modular ML pipeline (ingestion → training → evaluation)
- FastAPI backend + Streamlit UI
- ROUGE metric evaluation
Tech: Transformers, FastAPI, Streamlit, T5 Model
Deep learning CNN for medical image classification
🔥 Key Features:
- Transfer learning with pre-trained models
- DVC for experiment tracking & versioning
- Complete MLOps workflow
- Production-ready deployment pipeline
Tech: TensorFlow, Keras, DVC, Transfer Learning
End-to-end ML pipeline with interactive web interface
🔥 Key Features:
- Multiple ML models (Logistic Regression, KNN, SVM)
- Complete data preprocessing & feature engineering
- Flask API + React frontend
- Real-time predictions
Tech: Python, Scikit-learn, Flask, React
Fine-tuned LLM for explaining Dockerfiles
🔥 Key Features:
- Llama2 model fine-tuned with LoRA
- PEFT for efficient training
- Specialized for DevOps documentation
- Custom dataset for Dockerfile explanations
Tech: Llama2, Transformers, PEFT, LoRA
I'm always excited to collaborate on:
- 🧠 AI/ML Research Projects - Pushing boundaries in model performance
- 🤖 Multi-Agent Systems - Building intelligent automation
- 🚀 Production ML - Taking models from notebook to production
- 🌐 Open Source - Contributing to the community
Have an interesting project or just want to chat about AI?
