Назад
обновлено 14 часов назад

Data Scientist (Agentic Systems)

120 000 - 180 000$
Формат работы
remote (только USA)
Тип работы
fulltime
Грейд
senior
Английский
b2
Страна
US
Вакансия из списка Hirify.GlobalВакансия из Hirify RU Global, списка компаний с восточно-европейскими корнями
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/

TL;DR

Data Scientist (Agentic Systems): Building and optimizing next-generation agentic systems for cybersecurity with an accent on LLM post-training, reinforcement learning, and complex agent workflows. Focus on designing benchmarking evals, improving model reliability, and integrating AI agents into real-world security operations.

Location: Must be based in the USA

Salary: $120,000–$180,000 per year

Company

A global leader in cybersecurity providing an AI-native platform to stop breaches and protect modern organizations.

What you will do

  • Post-train LLMs and agents using supervised fine-tuning and reinforcement learning (RLHF/RLAIF, PPO/GRPO/DPO) to automate security analyst procedures.
  • Devise AI agents and complex workflows including planning loops, tool calling, and memory management.
  • Establish objective criteria for benchmarking agentic systems using LLM-as-judge pipelines and statistical rigor.
  • Collaborate with Engineering, Data Science, and Managed Services teams to transition research prototypes into production.
  • Optimize prompts and inference performance to maximize model utility.
  • Research and prototype state-of-the-art methods in agentic planning and reasoning.

Requirements

  • Must be based in the USA
  • PhD-level depth of understanding in modern machine learning research and experimental design.
  • Strong command of LLM training fundamentals including architecture, optimization, and scaling behavior.
  • Core skills in reinforcement learning and policy optimization (PPO/GRPO/DPO).
  • Experience building agentic systems with architectures like ReAct, planning, and retrieval.
  • Fluency with Python, PyTorch, and LLM training/serving stacks like Hugging Face, DeepSpeed, and vLLM.

Nice to have

  • Experience with synthetic data generation and task simulators.
  • Familiarity with inference-time scaling and verifier-guided decoding.
  • Knowledge of agent safety, guardrails, and failure analysis.
  • Background in cybersecurity or passion for applying ML to the security domain.

Culture & Benefits

  • Market-leading compensation and equity awards.
  • Comprehensive physical and mental wellness programs.
  • Competitive vacation and holiday policies.
  • Paid parental and adoption leaves.
  • Professional development opportunities and vibrant office culture.

Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →