class Rajveer:
name = "Rajveer Singhal"
role = ["Data Scientist", "ML Engineer", "AI Builder"]
education = "MCA โ Data Science"
location = "India ๐ฎ๐ณ"
building = "Analytix AI ๐ โ LLM-powered analytics for everyone"
tech_focus = {
"core" : ["Machine Learning", "Deep Learning", "NLP", "GenAI"],
"ops" : ["MLOps", "Model Deployment", "API Design"],
"data" : ["EDA", "Feature Engineering", "SQL", "Data Pipelines"],
}
current = "Fine-tuning LLMs + building production ML systems"
fun_fact = "I debug models at 2 AM โ and enjoy it ๐"
def philosophy(self):
return "Data without insight is just noise. I build the bridge. ๐"| Project | Description | Stack | Status |
|---|---|---|---|
| ๐ง Analytix AI | LLM-powered analytics platform โ ask questions, get insights, no SQL needed | FastAPI LangChain OpenAI React |
๐ข Active |
| ๐ฌ NLP Sentiment Engine | Real-time sentiment analysis REST API with 91% accuracy on custom dataset | Transformers Flask BERT |
โ Complete |
| ๐ ML Training Dashboard | Interactive UI to train, evaluate & visualize ML models without writing code | Streamlit Scikit-learn Plotly |
๐ง Improving |
| ๐ฏ Customer Churn Predictor | End-to-end churn prediction with SHAP explainability (89% accuracy) | XGBoost SHAP FastAPI |
โ Complete |
๐ฌ Fine-tuning open-source LLMs (Mistral, LLaMA)
โ๏ธ MLOps with MLflow + DVC + GitHub Actions
๐ RAG Pipelines & Vector Databases (Pinecone, Chroma)
โ๏ธ Cloud ML Deployment (AWS SageMaker / GCP Vertex AI)
- ๐ Building Analytix AI โ an LLM-powered analytics SaaS from scratch
- ๐ง Deployed 4+ ML models to production with REST API interfaces
- ๐ Built end-to-end pipelines handling 10K+ records with automated retraining
- ๐ฌ Achieved 91% accuracy on custom NLP sentiment classification task
- ๐ฏ Designed customer churn model reducing false negatives by ~23% via SHAP-driven feature tuning
๐ก Open to Data Science / ML Engineering roles โ full-time & internships. Feel free to reach out to discuss ideas, projects, or opportunities!

