ML foundations,
production experience

Ranjan Kumar Choubey

M.Tech in Data Science from ISI Kolkata (AIR 11), with a dissertation on reducing attention complexity in Graph Transformers. Built and deployed an electricity-trading forecast model at Tata Power and improved a payment-default model at Allstate.

At Deloitte, on a Walmart engagement, I build the Spark-Scala data products on GCP that feed Walmart's enterprise planning and forecasting. I also teach Probability, Statistics and Machine Learning for GATE DA, and build full-stack apps — for example an AI-powered notebook.

3+ Years Experience
AIR 11 ISI Entrance Exam
2 Forecast Models in Production
4x Star of the Month

Where I've worked

My journey across machine learning, production systems and teaching.

Consultant Deloitte
May 2025 – Present
Client: Walmart International (since Jul 2025)
  • Enterprise Planning: independently delivered the Sales module — Spark-Scala pipelines that build sales and margin data, aggregate it over rolling N-week windows into Hudi, and generate SAP BW planning feeds; validated with SAP and BW teams. Runs daily, feeding Walmart's daily planning.
  • Order substitutions: built and deployed the substitution analytics layer over Kafka order events — nil/short picks, substitution outcomes and fill rate.
  • Markdown reporting (current): leading discovery to deployment — source profiling, KPI definitions and validated BigQuery views.
  • Ingestion: brought 100+ SAP and Kafka source tables into the GCS datalake (Airflow, CI/CD).
  • Appreciated by the Walmart manager for the Enterprise Planning and substitution deliveries.
GATE Faculty (Part-time) IMS Academy
2025 – Present
  • Teach Data Structures to GATE CS aspirants, and Probability & Statistics and Machine Learning to GATE DA aspirants—helping students build strong conceptual foundations for their exams.
Private Tutor & Student Mentor Independent
Ongoing & Previous Mentorship
  • Provide personalized tutoring to B.Tech students, adapting explanations and guidance to each student's coursework and learning goals.
  • Previously taught and mentored school students preparing for Board examinations and Olympiads, helping them strengthen core concepts and problem-solving skills.
Data Science Intern Allstate India
May 2024 – Jul 2024
  • Improved a baseline XGBoost model for payment-default prediction through geographic feature engineering.
  • Analyzed geographic data to identify default trends; presented to Bengaluru & US teams, leading to 3 targeted model enhancements.
Lead Engineer Tata Power
Sep 2022 – Jul 2023
  • Built and deployed ARIMA and XGBoost forecasts for the Haldia power plant in day-ahead time blocks.
  • Generation model: real-time PI System sensor data and Tata Steel data (temperature, pressure, boilers online, coal moisture). Sales model: IEX market data.
  • Designed backend REST APIs using FastAPI and Node.js to serve data for 10+ Power BI dashboards.
Graduate Engineer Trainee Tata Power
Sep 2021 – Sep 2022
  • Deployed ML models on on-premise servers, managing full lifecycle from development to production with CI/CD.
  • Built email automation for employee onboarding — extracting, parsing, and updating DB, eliminating manual entry for 100+ employees/year.
  • Automated Power BI reporting for HR, operations, and engineering teams, replacing Excel-based workflows.

Academic background

Strong foundation in computer science and data science.

M.Tech — Computer Science (Data Science)
Indian Statistical Institute, Kolkata 2023 – 2025
AIR 11 in ISI Entrance AIR 1 EWS Category
B.Tech — Information Technology
BIT Sindri, Dhanbad 2017 – 2021
CGPA 8.89 — Rank 1 BITSAA Scholar '18 & '19

Tech stack

What I use at work, and what I'm building with now.

Machine Learning
PyTorch scikit-learn XGBoost Time-Series Forecasting Graph Neural Networks Transformers
Languages
Python Scala SQL Bash
Big Data
Apache Airflow Apache Spark Kafka ETL/ELT BigQuery PostgreSQL MySQL
Cloud & Storage
GCP Vertex AI GCS Dataproc Firebase
LLMs & GenAI
LLM APIs RAG Embeddings FAISS Fine-tuning (LoRA) AI Agents (LangGraph) Hugging Face LangChain
Backend & APIs
FastAPI Node.js REST APIs
MLOps & DevOps
Docker Kubernetes MLflow Git CI/CD Linux Power BI
Math & Statistics
Probability Statistics Linear Algebra Pandas / NumPy

Real projects, real impact

ML models, research, and the data systems behind them.

Tata Power

IEX Energy Trading Forecast

ARIMA and XGBoost forecasts for the Haldia power plant: generation from PI System sensor and Tata Steel data, sales from IEX market data, for day-ahead trading. Built and deployed.

Python XGBoost ARIMA FastAPI
ISI Kolkata / M.Tech Dissertation

Subgraph-Partitioned Graph Transformer

Reduced Graph Transformer attention from O(N²) to O(K²) by partitioning the graph into K subgraphs (METIS) and attending over a supergraph of GCN + Laplacian-embedding subgraph tokens.

PyTorch PyG METIS Research
Allstate India

Payment Default Prediction

Improved an XGBoost payment-default prediction model using geographic feature engineering and hyperparameter tuning.

XGBoost Python Feature Eng.
Deloitte / Walmart

Planning & Forecasting Data Products

Spark-Scala data products on GCP — sales and margin over rolling windows, order-substitution analytics — that feed Walmart's daily enterprise planning and forecasting.

Spark-Scala BigQuery Kafka GCP
Personal Project

Notebook App

AI-powered note-taking app with rich editor, code blocks, math rendering, drawing canvas, PDF export. Integrates Gemini and Groq for AI writing. Deployed as PWA.

Next.js React Firebase AI TypeScript
Tata Power

Automated Reporting System

End-to-end automated Power BI dashboards for HR, customer operations, and engineering teams. REST APIs with FastAPI and Node.js serving 10+ dashboards.

Power BI FastAPI Node.js REST APIs

Let's connect

Open to Machine Learning Engineer and ML Platform roles in Bengaluru.