ServiceNow is seeking a Principal Research Engineer/Scientist to independently lead a major workstream in agent learning and recursive self-improvement. The role involves designing and executing end-to-end research projects to improve long-horizon enterprise agents across areas including planning, reasoning, memory, tool use, retrieval, computer use, multi-agent coordination, and verification. The role researches both model post-training methods and harness-level optimization, and builds improvement flywheels that use trajectories and production-safe signals to identify failure modes, generate or curate data, test interventions, and measure generalization. It also involves creating realistic, stateful training environments and benchmarks for enterprise workflows, with programmatic verifiers and calibrated human or model-based graders where deterministic grading is not possible. The Principal Research Engineer/Scientist runs rigorous ablations and scaling experiments; develops capabilities across modalities and multilingual or cross-lingual settings; and builds reproducible distributed pipelines for training, rollout generation, evaluation, and inference. The position partners with research, engineering, and product teams to bring validated methods into reliable enterprise systems and communicates results through reviews, reports, publications, patents, open-source contributions, and recommendations.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
Job Type
Full-time
Career Level
Principal
Education Level
Ph.D. or professional degree