Headline

Data Engineer at Deloitte I MLOps & Predictive Analytics for FMCG | SQL | Python | Azure I VNIT’22

About

I’m Tanmay Kadlag — a Consultant at Deloitte who builds production ML systems for FMCG teams. I specialize in MLOps, predictive analytics, and data engineering (Scala & Python) to reduce manual effort and turn forecasts into operational actions. Key wins: Industrialized a country-scale predictive analytics pipeline and accelerated model deployment by 30% (integrated with production platforms). Built end-to-end Scala/Python frameworks for data prep that improved sales forecasting accuracy and model readiness. Automated reporting workflows to reduce manual work by 30% and cut Power BI refresh times by 25%. What I do now: lead cross-functional projects to operationalize ML models, set up model health & retraining triggers, and deliver dashboards & automation that front-line teams can act on. If you’re scaling ML for revenue or ops impact in FMCG (or nearby industries), let’s connect.

Location

Pune District

AI stack

codex

cursor

Work experience

Data Engineer at Deloitte, May 2024 - Present · 2 yrs 4 mos, Pune District, Maharashtra, India, Industrialized a country-scale predictive analytics pipeline and reduced model deployment time by 30%, enabling faster model releases and rollbacks. Designed a reusable Scala + Python data-generation framework that increased feature coverage and improved sales-forecast accuracy. Launched a new ML model (clustering + supervised ensemble) that improved target performance by 10% and was integrated into operational workflows. Implemented automated model-health monitoring and retraining triggers, increasing pipeline reliability and reducing manual interventions by an estimated 20%.

Analyst | Cloud Engineering at Deloitte, Jul 2022 - May 2024 · 1 yr 11 mos, Mumbai, Maharashtra, India, Increased predictive-model accuracy by 15% via improved preprocessing and feature engineering for churn and sales-recommendation systems. Automated report generation with Scala, cutting manual reporting effort by 30% and enabling near-real-time performance tracking. Optimized Power BI backends to reduce refresh times by 25%, improving visibility for sales and analytics teams. Standardized data-cleaning pipelines, improving data readiness for modeling by 40%.

Education

Bachelor of Technology - BTech, Electrical and Electronics Engineering, Visvesvaraya National Institute of Technology

Public links

github: https://github.com/takadlag