Data Scientist

This is a fully remote position, open to applicants in United States.

📋 Description

• Engage at the nexus of Artificial Intelligence and Threat Research.

• Collaborate intimately with cybersecurity subject-matter experts to gain insights into analyst workflows and their security operations protocols.

• Conduct post-training of LLMs and agents through supervised fine-tuning and reinforcement learning (RLHF/RLAIF, PPO/GRPO/DPO, reward modeling) to automate analyst procedures and enhance reliability on actual security tasks.

• Develop AI agents and integrate them into progressively intricate workflows, including planning and reasoning loops, tool and function invocation, and retrieval and memory management.

• Investigate innovative methods for agentic planning and prototype cutting-edge techniques from existing literature.

• Set objective benchmarks for evaluating agentic systems, including evaluations, LLM-as-judge pipelines, and trajectory-level metrics, ensuring statistical rigor.

• Refine prompts and inference methods to maximize the potential of each model.

• Collaborate and coordinate with Engineering, Data Science, and Managed Services teams, working alongside engineers to advance prototypes to production.

• Stay abreast of advancements in the field of Artificial Intelligence and assist in identifying, defining, and prioritizing research areas.

⛳️ Requirements

• Strong foundational knowledge in machine learning, probability, and statistics, with an intuitive grasp of uncertainty, statistical skew/variance, and experimental design.

• PhD-level understanding in contemporary machine learning research; while a doctorate is not mandatory, equivalent expertise is expected, including the ability to read, critique, implement, and enhance existing papers.

• Proven experience in training generative models, with a solid grasp of LLM training fundamentals (architecture, optimization, tokenization, data, and scaling behavior).

• Reinforcement learning and post-training as a core competency: RLHF/RLAIF, policy optimization (PPO/GRPO/DPO), reward modeling, and constructing RL environments for agents.

• Demonstrated experience in building agentic systems, including agent architectures (ReAct, planning, reflection), tool and function invocation, and retrieval/memory/context management.

• Familiarity with systematic prompt optimization and designing and constructing evaluations for LLM systems.

• Proficient with GPUs, PyTorch, and the standard LLM training and serving stack (e.g., Hugging Face Transformers/TRL/PEFT, DeepSpeed/FSDP, vLLM/TGI/SGLang).

• Strong, reproducible research engineering capabilities: clean Python coding and disciplined experiment tracking that allows collaborators to build upon.

• Ability to independently navigate ambiguous and complex objectives, with clear communication within a large project team.

🏝️ Benefits

• Leading market compensation and equity awards.

• Comprehensive programs for physical and mental wellness.

• Competitive vacation and holiday policies to promote recharging.

• Paid parental and adoption leave.

• Professional development opportunities available for all employees, regardless of level or role.

• Employee Networks, local community groups, and volunteer opportunities to foster connections.

• Dynamic office culture featuring world-class amenities.

• Recognized as a Great Place to Work Certified™ globally.