4 дня назад
Staff Software Engineer (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Staff Software Engineer (AI) (LLM agents/full stack): Building Otto, an AI agent platform that reasons over customer context and takes secure actions across payroll, benefits, and compliance systems with an accent on agent runtimes, tool layers, Rails/GraphQL services, and advocate-facing experiences. Focus on designing reliable evaluation and observability systems, controlling the blast radius of actions on live accounts, and improving agent behavior from real interactions.
Location: Denver, CO; New York, NY; or San Francisco, CA, United States. Employees based in these locations are expected to work from the office approximately 2–3 days per week. A secure, reliable, and consistent internet connection is required when working outside a office.
Company
provides payroll, health insurance, 401(k), and HR services to more than 500,000 small businesses.
What you will do
- Design and build the Otto AI agent platform, including the runtime, agent graph, tool and action layer, Rails/GraphQL backend, and advocate-facing product experience.
- Expand secure agent capabilities with autonomous tool-using actions, human approval workflows, and auditable identity and permissions models.
- Build evaluation and observability systems using LLM-based judges, deterministic scorers, offline gates, and production monitoring.
- Work with LLM prompting, tool and function design, retrieval, and feedback loops based on advocate interactions.
- Partner with Product, Data, Design, and Customer Experience teams to deliver capabilities that address customer and advocate problems.
- Mentor engineers and establish responsible practices for building agentic systems at scale.
Requirements
- Typically 8+ years of software engineering experience with end-to-end ownership of systems.
- Production experience building and shipping LLM-powered or agentic systems, including tool-calling agents, retrieval, or evaluation systems.
- Strong full-stack range across backend services, LLM runtimes, and customer-facing products.
- Experience owning multi-quarter projects with measurable user impact at scale.
- Sound judgment regarding safety and blast radius when software takes actions on real customer accounts.
- Ability to mentor engineers and raise engineering standards.
Nice to have
- Experience with agent frameworks, LLM evaluations, LLM-as-judge systems, retrieval, or MCP-style tool architectures.
- Experience with Ruby on Rails, JavaScript, or React, or the ability to ramp up quickly.
- Familiarity with GraphQL and federated backends.
- Experience in collaborative, test-driven, high-autonomy environments with clear stakeholder communication.
Culture & Benefits
- Full-time employees receive competitive base pay, benefits, and RSU equity.
- AI tools are integrated into how work gets done, with an expectation that employees develop relevant AI fluency.
- Work is organized around customer outcomes and a short path from idea to production.
- supports employees from diverse backgrounds and provides reasonable accommodations during the hiring process.
- Hybrid employees must maintain a secure and reliable internet connection when working outside the office.
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 →