Data Analyst · Calicut, Kerala
I turn multi-table transaction logs into dashboards your team actually checks before making a call. SQL to insight, without the three-day wait.
Stockouts reduced
Faster exec decisions
Attrition surfaced
ML model accuracy
About
I'm a Computer Science graduate working as a Data Science Intern at Techolas Technologies in Calicut, where I spend most days moving between raw SQL tables and the Power BI dashboards that store managers and HR teams open every morning.
My habit is to look for the anomaly before the chart: a broken foreign key, a duplicated transaction, a store that's quietly running out of stock. Fixing that first is usually what makes the dashboard worth checking twice.
Outside internship work, I build end-to-end projects — from a star-schema retail model to an income-prediction ML app — to keep my SQL, Python, and visualization skills sharp against real, messy data rather than tutorials.
Featured work
Power BI · DAX · SQL
Multi-store inventory data lived in disconnected tables, so stockouts were usually caught after shelves were already empty.
Rebuilt the data model on a clean star schema and shipped conditional-DAX alerts that cut critical stockouts by 18% and cut store-manager check time by 35%.
Tableau · Python
Leadership suspected attrition was rising but had no view into where or why people were actually leaving.
Analyzed 1,470 employee records to surface a 16.12% attrition rate, pinpointing that 47.3% of exits clustered in the 25–34 age bracket to focus retention spend.
Python · Power BI
Regional revenue trends and shipping delays were reviewed manually every month, a day of work per cycle.
Built a Pandas/NumPy pipeline feeding an interactive Power BI dashboard, isolating shipping bottlenecks that improved delivery efficiency by 12%.
Python · Scikit-learn · Streamlit
Benchmarked four classification models on census data, reaching 85–87% accuracy, then shipped a live Streamlit app returning predictions in under 2 seconds.
View repoExperience
Techolas Technologies, Calicut
Aug 2025 — Jun 2026 · 11 mos
Redesigned the data model behind multi-store retail data using a star schema, cutting broken dimension links by 25% and restoring trust in the numbers.
Shipped 4 interactive KPI dashboards across e-commerce and HR by writing SQL/MySQL queries and automated ETL pipelines, cutting the executive reporting cycle by 40%.
Built conditional-DAX (SWITCH(TRUE())) alerting in Power BI and Tableau that flagged low-stock items early, reducing critical stockouts by 18% across store locations.
Audited 30,000+ operational records with Python (Pandas, NumPy) and macros, removing duplicate entries so downstream reports stayed accurate on a fixed schedule.
Skills
Certifications
Contact
Open to full-time Data Analyst roles across Calicut, Kannur, Vadakara, and Thalassery — and to remote/hybrid opportunities anywhere else. I reply within a day.
Email me directly