Ujjwal Chaurasia — AI/ML Engineer

About

A calm mind behind the model.

I'm Ujjwal Chaurasia, an AI/ML undergraduate at Vivekananda Institute of Professional Studies who enjoys turning messy data into clear, usable systems. My internship experience spans churn prediction, sales forecasting dashboards, and LLM post-training evaluation — from raw data to a working, deployable result.

Outside of coursework in NLP, image processing, and pattern recognition, I spend time refining projects on GitHub and tuning my own Linux setup — small, deliberate improvements, the same philosophy I bring to model building.

1

Flagship Project

2

Internships

0.848

Best ROC-AUC

Career

Experience

LLM Post-Training Intern

Apr 2026 – Jul 2026

Ethara AI

Improved model response quality across assigned evaluation batches by auditing outputs for accuracy, coherence, and guideline adherence using structured rubrics and iterative feedback from Quality Leads.
Strengthened evaluation consistency across the team by documenting recurring edge cases and feedback patterns, applying industry-standard AI data evaluation workflows under senior mentorship.

LLM EvaluationRubricsNLP

Machine Learning Intern

Jun 2025 – Jul 2025

Future Interns

Built and benchmarked Logistic Regression, Random Forest, and XGBoost churn prediction models on a 7,043-customer dataset — selected XGBoost as top performer with 80.5% accuracy and 0.848 ROC-AUC.
Built a national sales forecasting dashboard in Power BI covering 4 regions and a multi-year timeline (2014–2018) to support inventory and demand-planning decisions, using SQL to query and transform historical sales data.
Delivered end-to-end ML pipelines spanning forecasting, classification, and NLP use cases, using Git/GitHub for version control and collaborative development.

PythonSQLPower BIXGBoost

Education

Bachelor of Technology, Artificial Intelligence & Machine Learning

Vivekananda Institute of Professional Studies, New Delhi

2023 – Present

Toolkit

Skills & Technologies

Languages & Tools

Python

SQL

Git & GitHub

Machine Learning

XGBoost

Random Forest

Logistic Regression

Model Evaluation

Foundations

Data Structures & Algorithms

Natural Language Processing

Data Visualization

Pandas

NumPy

Power BI

scikit-learn

Selected Work

Projects

Customer Churn Prediction Dashboard

ProblemIdentify which of 7,043 customers were likely to churn, so retention efforts could be prioritized.

ApproachBuilt and benchmarked Logistic Regression, Random Forest, and XGBoost models, comparing precision, recall, and F1-score.

OutcomeXGBoost selected as top performer — 80.5% accuracy, 0.848 ROC-AUC.

Pythonscikit-learnXGBoost

View Case→

Get in Touch

Let's build something.

Open to Machine Learning Engineer and Python Developer internship or entry-level roles. Reach out — I usually reply within a day.

Say Hello View / Download Resume

Emailujjwalkumarchaurasiya@gmail.com GitHubgithub.com/uctheinevitable LinkedInlinkedin.com/in/ujjwal-chaurasia

StatusAvailable for internships