Aalokita Chibb

Data Analyst / Data Scientist

Chandigarh, India 85 · Excellent

About

Final-year B.E. CSE (Big Data Analytics) student with hands-on experience building ML pipelines, time-series forecasting systems, and analytics dashboards on real-world datasets. Delivered a demand forecasting model across 42,000+ time series and an ML-driven risk detection system adopted by a university CS department. Google Data Analytics certified and AWS Cloud Practitioner; passionate about translating messy data into decisions that drive measurable business outcomes.

Skills & Expertise (48)

Python Intermediate

7.0/10

Years Exp

Power BI Intermediate

6.5/10

Years Exp

scikit-learn Intermediate

6.5/10

Years Exp

SQL Intermediate

6.0/10

Years Exp

Git Intermediate

6.0/10

Years Exp

LightGBM Intermediate

6.0/10

Years Exp

XGBoost Intermediate

6.0/10

Years Exp

Tableau Intermediate

5.5/10

Years Exp

Apache Spark Intermediate

5.0/10

Years Exp

Docker Intermediate

5.0/10

Years Exp

AWS (S3 SQLite Vercel AWS (Glue AWS (Athena AWS (Lambda AWS (EC2 AWS (IAM MongoDb LangChain LlamaIndex RAG CrewAI Gemini API Pinecone ChromaDB MLflow FastAPI K-Means Streamlit Plotly Matplotlib Seaborn Scala Shell Scripting Prophet SHAP Random Forest MySql Ensemble Methods DBT Airflow Hadoop Hive Kafka ETL Design Data Modeling Postgresql

Work Experience

HR Attrition Analytics & Prediction

IBM

Jan 2025 - Jan 2025

Building an attrition prediction model on the IBM HR Analytics dataset (1,400+ employees) targeting AUC-ROC > 0.85. Utilized SHAP explainability frameworks to identify top drivers of turnover by department and tenure band. Conducted cohort analysis by hire year, job role, and salary band; built Power BI executive dashboard with drill-through maps.

Supply Chain Demand Forecasting with LLM Copilot

Walmart

Jan 2024 - Jan 2024

Built a time-series forecasting pipeline on the Walmart M5 dataset (42,000+ time series). Engineered 40+ lag, rolling-window, and calendar features via dbt, reducing manual feature prep time by 60%. Developed a multi-model ensemble (Prophet + LightGBM + TFT) improving MAPE by 15% over baseline; evaluated on RMSSE, MAE, and WRMSSE. Integrated a LangChain NLP copilot for natural-language anomaly querying, stockout risks, and promotional lift quantification.

KACHUA Student Risk Analytics Platform

Chandigarh University

Jan 2024 - Jan 2024

Adopted by Chandigarh University CS department; 71-feature behavioral engine flags at-risk students with explainable interventions. Engineered end-to-end pipeline: survey ingestion → feature extraction → ML inference → LLM counseling reports. Key baseline insight: students below 75% class attendance showed a 42% higher failure rate. Built a teacher analytics dashboard with bulk-upload batch assessment and cohort-level risk dashboards.