The opportunity
At FourKites we have the opportunity to tackle complex challenges with real-world impacts. Whether it's medical supplies from Cardinal Health or groceries for Walmart, the FourKites platform helps customers operate global supply chains that are efficient, agile and sustainable.
What you'll do
Design, build, and productionize ML models for problems like ETA/ATA: prediction, using regression, classification, and time-series forecasting techniques
Develop NLP/LLM-based extraction pipelines for message-based ETA and status: updates (text extraction, entity recognition)
Own models end-to-end: data pipeline → training → deployment → monitoring → retraining
Work with noisy, real-world logistics and supply chain data (GPS pings, check: calls, carrier data) rather than clean, pre-processed datasets
Diagnose gaps between offline evaluation performance and live production accuracy, and drive fixes
Build and maintain automated training/retraining pipelines using orchestration tools such as Airflow
What they're looking for
- Set up and maintain model monitoring and observability (e.g., Grafana) to: catch drift and degradation proactively
- Replace manual or rule-based processes with ML-driven automation (e.g., automating manual check calls)
- Translate model performance improvements into business impact: operational savings, efficiency gains, and deal-relevant outcomes
- Mentor and guide other data scientists/engineers on technical approach and best practices
