Hello, I am

Rohit Saroj

Experienced Data Scientist skilled in Python, SQL, Generative AI, AWS, and advanced data analysis. I specialize in automating processes, designing AI/ML solutions, and deploying production-ready models to solve complex business challenges.

Get in Touch Download CV

Professional Experience

Aug 2025 – Present

Sr. Data Scientist

Teamlease (Client: Deloitte USI) | Gurgaon, IN

  • Developing a Generative AI–powered chatbot (RWE Agent) using LangGraph for automated clinical cohort building.
  • Implemented prompt optimization using DSPy and integrated Claude 3.5 via VOX API.
  • Designed a human-in-the-loop workflow in LangGraph for user review of extracted criteria.
  • Built an AWS Knowledge Base mapping system to link criteria with Snowflake schema.
  • Engineered a study recommendation module leveraging AWS KB and K-Means clustering.
  • Designed automated ETL pipelines to fetch and preprocess study metadata for S3/AWS KB ingestion.
LangGraph DSPy AWS Bedrock Snowflake FastAPI
Jun 2023 – April 2025

Data Scientist

Stem Inc | Gurgaon, IN

  • Designed and deployed a RAG-based application using OpenAI embeddings & Pinecone, reducing resolution time by 40%.
  • Enhanced predictive accuracy of site load forecasting models, increasing reliability.
  • Automated analytics workflows in Jupyter Notebook, boosting triaging efficiency by 25%.
  • Leveraged AWS, Jenkins, and CI/CD pipelines to deploy scalable data solutions.
Python LangChain OpenAI API Pinecone Jenkins
Sep 2022 – Jun 2023

Analyst

Manikaran Analytics Limited | New Delhi, IN

  • Refined wind energy forecasting models, reducing Mean Absolute Error (MAE) by 12%.
  • Automated recurring analysis tasks using Python scripts, saving 25% of team resources.
  • Built wind speed prediction models with Random Forest and XGBoost, enhancing precision by 15%.
XGBoost Random Forest Pandas SQL
Mar 2020 – Aug 2022

Performance Analyst

Emergya Wind Turbines Pvt Ltd | Chennai, IN

  • Designed a CNN-based AI model to classify wind turbine power curves, improving efficiency by 50%.
  • Predicted component failures using ML algorithms, reducing maintenance costs by 20%.
  • Automated data reporting via Python-based GUIs.
TensorFlow CNN PySpark Power BI
Mar 2017 – Mar 2020

Engineer

WIND WORLD INDIA LTD | Mumbai, IN

  • Boosted energy yield by 5% through in-depth turbine performance analysis.
  • Conducted root cause analyses to identify and rectify underperformance issues.
Excel SQL Python

Technical Skills

Programming & Tools

Python SQL Git Docker Jenkins Jira Bitbucket

AI & Machine Learning

Generative AI LLMs LangChain NLP Deep Learning TensorFlow PyTorch

Data & Analytics

Pandas NumPy Scikit-learn EDA Feature Engineering Power BI Plotly

Frameworks & Cloud

FastAPI Flask Streamlit PySpark AWS (S3, EC2, Bedrock)

Education & Certifications

Education

  • PGDBM Operation Management

    NMIMS, Mumbai | Jul 2022

  • B.E. Instrumentation Engineering

    Rajiv Gandhi Institute of Technology, Mumbai | Aug 2016

Certifications

  • Prompt Design in Vertex AI - Google
  • SQL for Data Science – Great Learning
  • The Data Science Course 2022 – Udemy
  • Generative AI with Langchain – Udemy

Get In Touch

I'm currently open to new opportunities. Whether you have a question or just want to say hi, I'll try my best to get back to you!

Email

rohitsaroj29@gmail.com

LinkedIn

Connect with me

Medium

Rohit (1.5K+ followers)

Rohit's Assistant

Online
Hi! I'm Rohit's virtual assistant. Ask me about his experience, skills, or how to get in touch!