SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
FourKites is an AI-driven supply chain visibility platform that processes over 3.2 million supply chain events daily for 1,600+ global brands. As a Senior Data Scientist, you will design, build, and own machine learning models that power core prediction problems across the platform, including ETA/ATA forecasting and message-based status extraction.
You will work end-to-end from data pipeline to production deployment and monitoring, transforming noisy real-world logistics data into models that run at scale and directly impact customer outcomes. Key responsibilities include designing and productionizing ML models using regression, classification, and time-series forecasting techniques; developing NLP/LLM-based extraction pipelines for message-based ETA and status updates; owning models through their full lifecycle including data pipeline, training, deployment, monitoring, and retraining; working with real-world logistics and supply chain data (GPS pings, check calls, carrier data); diagnosing gaps between offline evaluation and live production accuracy; building automated training/retraining pipelines using Airflow; setting up model monitoring and observability using tools like Grafana; replacing manual or rule-based processes with ML-driven automation; translating model improvements into business impact; and mentoring other data scientists on technical approach and best practices.
You should have strong ML fundamentals across regression, classification, and time-series forecasting; NLP experience with text extraction, entity recognition, or LLM-based extraction; proven production ML experience shipping models serving real traffic; strong Python and SQL skills with pandas, scikit-learn, and experience querying large datasets (Redshift/Snowflake preferred); cloud and data infrastructure experience with AWS (S3, EC2) and orchestration tools like Airflow; experience with model monitoring and observability tooling; comfort working with noisy, real-world data; ability to diagnose and close gaps between offline and live performance; track record of replacing manual processes with ML solutions; ability to translate model output into business value; cross-functional collaboration experience; mentoring experience; and ability to make build-vs-buy and architecture tradeoffs independently. Nice-to-have skills include logistics/supply chain/transportation experience, familiarity with streaming data (Kafka), and exposure to LLM/GenAI applications in production.
