Headline

Data Scientist | Machine Learning | GenAI & LLMs | Customer Analytics | Forecasting & Optimization | Python, SQL

About

I’m a Data Scientist with 4+ years of experience building machine learning and AI solutions across customer analytics, predictive modeling, forecasting, optimization, and Generative AI. I work across the ML lifecycle—from problem definition and data analysis to feature engineering, model development, evaluation, and production deployment. A few examples of my work: • Built a production optimization engine that generated ~$2M in annual transportation savings using Mixed Integer Linear Programming (MILP) and Gurobi. • Developed an ensemble demand forecasting system using XGBoost and time-series methods, contributing ~$1M in annual operational savings. • Built customer segmentation and propensity models across 50,000+ retail outlets, increasing targeted campaign conversion from 15% to 22%. • Built and deployed a GPT-powered RAG platform using LangChain, embeddings, and prompt engineering, reducing analyst response time by ~60%. I’m particularly interested in problems involving prediction, customer analytics, personalization, experimentation, and AI-powered products. I enjoy working on problems where the goal is not just to build a model, but to define the right objective, features, evaluation metrics, and business outcome. My areas of focus include Machine Learning, customer analytics, predictive modeling, forecasting, optimization, Generative AI, LLM applications, and RAG. Skills: Python • SQL • Machine Learning • Statistical Modeling • Predictive Analytics • XGBoost • LightGBM • Deep Learning • Time Series Forecasting • A/B Testing & Experimentation • Customer Segmentation • Propensity Modeling • Optimization • NLP • Generative AI • LLMs • RAG • MLOps I’m interested in Data Science and Applied ML opportunities where I can work on high-impact products, own problems end-to-end, and build ML systems that reach real users. 📩 Contact: ayusht474@gmail.com

Location

Bengaluru

AI stack

codex

github copilot

langchain

openai

Work experience

Data Scientist | Machine Learning | Customer Analytics | Optimization | GenAI at AB InBev GCC India, Jul 2025 - Present · 1 yr 4 mos, Bengaluru, Karnataka, India · Hybrid, Built a retail outlet segmentation and propensity modeling solution for 5,000+ outlets to identify high-value customers and improve campaign targeting. • Used SQL for data preparation and analytical feature creation across customer transaction and engagement data. • Engineered RFM, product mix, purchase frequency, and engagement features to capture outlet behavior and commercial potential. • Applied K-Means clustering to identify distinct outlet segments and developed a LightGBM classification model to rank outlets based on propensity for targeted campaigns. • Increased campaign conversion from 15% to 22% by enabling sales teams to prioritize higher-propensity outlets. • Partnered with business and sales stakeholders to translate customer behavior into actionable segments, propensity scores, and targeted sales strategies.

Junior Data Scientist at AB InBev GCC India, Jun 2024 - Jul 2025 · 1 yr 2 mos, Bengaluru, Karnataka, India · Hybrid, Worked on machine learning, forecasting, and optimization solutions for supply chain planning, with a focus on improving transportation, reusable packaging, and inventory decisions. • Delivered ~USD 2M in annual transportation cost savings by developing a zone-level Mixed Integer Linear Programming (MILP) optimization model in Gurobi to optimize brewery-to-distribution center shipment allocation and customer fulfillment decisions across a network of 1,000+ nodes. • Developed an ensemble demand forecasting framework combining XGBoost with ARIMA and Prophet to predict empty bottle returns across 500+ SKU–distribution center combinations. • Engineered forecasting features from historical return patterns, inventory levels, turnaround cycles, and seasonality, achieving approximately 10% MAPE on high-volume SKUs. • Forecast outputs supported improved empties allocation and supply chain planning, contributing approximately $1M in annual savings.

Data Scientist at Deloitte, Jun 2022 - Jun 2024 · 2 yrs 1 mo, Bengaluru, Karnataka, India, • Engineered an anomaly detection and triage pipeline over 5M+ weekly insurance claims using regression residual analysis to identify abnormal provider KPIs and rank high-risk cases for analyst review, achieving 78% validation on recommended cases and reducing detection latency by several hours. • Built and deployed a production RAG application for a Fortune 500 healthcare client using LangChain, OpenAI embeddings, Azure AI Search, and GPT-3.5, implementing semantic retrieval and metadata filtering across 5,000+ enterprise documents. • Automated 65% of user queries through the RAG system, reducing analyst workload and improving access to information across enterprise documents.

Education

Master of Science - MS (Executive), Information Science- Machine Learning, University of Arizona

Bachelor of Technology - BTech, Information Technology, Indian institute of information technology una