About this role
Design and build agentic experimentation layers and evaluation harnesses to benchmark model and agent performance, integrate LLM-based and foundation models for time-series forecasting (ETA and demand), and implement operational safeguards for autonomous experimentation systems while partnering with OR and ML engineering teams.
Skills for this role
Evaluation harness designBenchmarkingAgent-based systemsAgentic systemsExperimentation frameworksLLM-based/foundation modelsTime-series forecastingPythonPyTorchHugging FaceCommunicationOptimization/OR collaborationOperational safeguardsFoundation model fine-tuningPost-training techniquesReinforcement LearningRLHFRetrieval-Augmented Generation (RAG)ETA predictionDemand forecastingAutomated regression checksHuman-in-the-loop systemsAutomated experimentationLogistics domain knowledgeChronos-style models