Sathwik Krishna Lanke

Developed and productionized a Python + Scikit-learn churn prediction model to reduce monthly rider attrition by 2,000 users; engineered cloud-native pipelines with FastAPI, Apache Kafka, and AWS Lambda to process 50K+ ride events/sec; integrated computer vision (DETR, Grounding DINO) into traffic management workflows; architected microservice components and optimized CI/CD pipelines; mentored engineers, reducing release cycles and improving reliability.

Developed and productionized a Python + Scikit-learn churn prediction model to reduce monthly rider attrition by 2,000 users and strengthen retention. Engineered cloud-native pipelines with FastAPI, Apache Kafka, and AWS Lambda to process 50K+ ride events/sec and improve ETA accuracy across 3.2M trips annually. Directed the integration of computer vision algorithms (DETR, Grounding DINO) into Uber's traffic management workflows, reducing hazard response times by 4+ days. Architected microservice components and optimized CI/CD pipelines, mentoring 3 engineers on data pipeline design and ML integration, cutting release cycles from 2 days to 6 hours.

Developed and productionized a Python + Scikit-learn churn prediction model to reduce monthly rider attrition by 2,000 users and strengthen retention strategy. Engineered cloud-native pipelines with FastAPI, Apache Kafka, and AWS Lambda to process 50K+ ride events/sec and improve ETA accuracy across 3.2M trips annually. Directed the integration of computer vision algorithms (DETR, Grounding DINO) into traffic management workflows, cutting hazard response times by 4+ days. Architected microservice components and optimized CI/CD pipelines, while mentoring 3 engineers on data pipeline design and ML integration, cutting release cycles from 2 days to 6 hours.