I'm Madanala Abhishek Varma, a B.Tech CSE graduate (May 2026) from Andhra Pradesh, India — currently a Data Analyst Intern at Bluestock Fintech, building end-to-end data pipelines on real financial data, while continuing to build in AI/ML and GenAI.
I started in frontend development (React) before moving into ML and now GenAI — I like understanding a system end-to-end rather than just one layer of it.
abhishek = {
"currently_doing" : "Data Analyst Intern @ Bluestock Fintech",
"currently_learning" : ["Agentic AI (LangGraph, CrewAI, AutoGen)", "RAG evaluation"],
"building" : "AI apps and data pipelines with real-world utility",
"open_to" : ["Data Analyst", "Junior ML Engineer", "GenAI Developer"],
"location" : "Andhra Pradesh, India 🇮🇳 (Remote-friendly)"
}
End-to-end ETL + analytics pipeline on real financial data
- Built a full pipeline on 87K+ real AMFI India NAV records, 40 fund schemes
- SQLite star schema, EDA with 15+ charts, risk-adjusted metrics — Sharpe, Sortino, Alpha, Beta, VaR
- Shipped a 4-page interactive Power BI dashboard with drill-through and cross-filtering
Rule-based classification, extraction, priority scoring, and a semantic assistant — over 1,080 messages
- Classifies messages into 6 categories, extracts tasks/events, detects and masks sensitive info
- L2 adds a multi-signal priority engine, related-message grouping, and a TF-IDF semantic QA assistant with privacy-aware routing
- Every decision is explainable — no black-box classification, full reasoning traced per output
- Predicts disease from 134 symptom-severity combinations across 41 disease categories
- Random Forest + Keras, trained on 4,920+ samples, 99% accuracy
- Interactive Streamlit web app, live and deployed
- Multi-task GenAI app — Summarize, Translate, Explain, Generate Email, Rewrite
- Dual LLM backend: Google Gemini (zero-cost) + OpenAI (BYOK), with seamless model switching
- Live on Streamlit
- 💼 Data Analyst Intern @ Bluestock Fintech — real fintech data pipelines
- 📚 Learning Agentic AI — LangGraph, CrewAI, AutoGen
- 📝 Applying for Data Analyst / Junior ML / GenAI roles
- 💡 Practicing SQL + Python (LeetCode) for interviews
From React components to RAG pipelines to ETL dashboards — still just connecting the dots.