Open to opportunities

Kartheek Garapati

@kartheekgarapati

AI/ML Engineer building end-to-end production models—anomaly detection, forecasting, and RAG—to improve decisions and customer experiences.

United States

Message

What I'm looking for

I’m looking for a role where I can build and ship production ML/AI end to end—anomaly detection, forecasting, and RAG included—while improving operational efficiency, decision-making, and customer outcomes with strong MLOps and monitoring practices.

I’m an AI/ML Engineer with 4 years of experience delivering machine learning, analytics, and AI applications across financial services, retail, and telecommunications. I build solutions end to end—from understanding business requirements and preparing data to developing models, deploying applications, and supporting production workflows.

At Fiserv, I developed Spark Structured Streaming pipelines to reduce investigation data availability to under 5 minutes, and engineered merchant behavioral features that power anomaly detection and case-prioritization. I also built RAG pipelines with Azure OpenAI GPT-4o, improving first-attempt retrieval accuracy from 69% to 88% while cutting document search time from 12 minutes to less than 2 minutes; I reduced false-positive alerts by ~20% and improved high-risk case precision from 0.62 to 0.77.

Previously at Tesco and Aricent, I accelerated data preparation and reporting, built forecasting and promotion-response models, and supported operational analytics. I’ve also validated a fault-resilient predictive maintenance approach in a production-style pipeline, using robust sensor degradation modeling, health scoring, and a robustness layer to significantly improve RUL prediction error.

Experience

Work history, roles, and key accomplishments

Current

ML Engineer

Current

Fiserv

Aug 2025 - Present (11 months)

Developed Spark Structured Streaming pipelines and Delta Lake datasets for payment operations, reducing investigation data availability from hourly batches to under 5 minutes. Built merchant behavioral features and anomaly detection workflows to reduce false-positive alerts by ~20%, and implemented RAG pipelines using Azure OpenAI GPT-4o with Pinecone; orchestrated training, batch scoring, and mon

Apache Spark Spark Structured Streaming Kafka Delta Lake Pyspark SQL Feature Engineering Anomaly Detection XGBoost Retrieval Augmented Generation (RAG)GPT 4o Pinecone S3 Cloudwatch Model Monitoring

Data Scientist

Tesco

Apr 2023 - Aug 2024 (1 year 4 months)

Built PySpark and SQL pipelines for promotion effectiveness and customer analytics, reducing campaign data preparation time from several hours to under 30 minutes. Engineered loyalty-based behavioral features, delivered LightGBM demand forecasting and XGBoost promotion-response models (improving forecast accuracy from 81% to 89%), and automated measurement/reporting using Python, Airflow, BigQuery

Pyspark SQL Data Engineering Feature Engineering Customer Analytics Demand Forecasting LightGBM XGBoost Python Airflow BigQuery Vertex AI Looker Data Modeling

Python Developer

Aricent

Jan 2022 - Mar 2023 (1 year 2 months)

Developed FastAPI-based services integrating OSS/BSS platforms and operational databases to ingest 2M+ network alarms, performance counters, and events per day. Built Python ETL and SQL processing services, implemented PostgreSQL and MongoDB data stores for monitoring dashboards, and created anomaly/analytics workflows to reduce investigation time from ~45 minutes to under 15 minutes; containerize