Data Scientist, Agentic Systems (Remote)
Core
Building next-generation agentic systems for cybersecurity by training LLMs, developing AI agents, and rigorously measuring their performance on real security tasks.
Role type
Senior IC machine-learning researcher (agentic systems)
Builds
AI agents, agentic workflows, and production-ready ML models for threat research and analyst automation
Domain
Cybersecurity + Generative AI / Reinforcement Learning
Deliverable
production ML models
Required skills
modern machine learning research, generative model training, reinforcement learning (RLHF/RLAIF, PPO/GRPO/DPO), agentic system architecture, systematic prompt optimization, LLM evaluation design, Python research engineering, GPU/PyTorch stack fluency
Preferred skills
synthetic data generation, inference-time scaling, agent safety/guardrails, model interpretability, open-source contributions
Technologies
PyTorch, Hugging Face Transformers/TRL/PEFT, DeepSpeed/FSDP, vLLM/TGI/SGLang, GPUs
Responsibilities
Post-train LLMs and agents using supervised fine-tuning and reinforcement learning; devise and combine AI agents into complex workflows with planning and reasoning loops; research new approaches to agentic planning and prototype state-of-the-art methods; establish objective criteria for benchmarking agentic systems; optimize prompts and inference; collaborate with engineering and managed services teams to take prototypes to production
Seniority
Senior, hands-on IC
