Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
развернуть
$160к-180к/год
$170к$190кмедиана$240к
По похожим ролям (49) ~ $190к/год
На СНГ рынке: $4.3к/мес ($3.6к - $5к)
88
Крутая вакансия
развернуть
У роли четкие задачи для senior-инженера, понятные этапы на первые месяцы и работа с современными AI-агентами в production; главное ограничение - работа доступна только в США.
Только для США
Передовые AI-технологии
Понятная зона ответственности
Влияние на бизнес
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Senior Engineer, AI/ML (AWS Bedrock AgentCore/Python): Building production agentic AI systems, tool interfaces, evaluation harnesses, and observability for a multi-tenant automotive SaaS platform with an accent on safe execution, production integrations, and business-aware engineering judgment. Focus on designing agent workflows, MCP servers, LLM-as-judge CI gates, guardrails, cost and latency controls, and reliable operation against real dealer data and transactions.
Location: Denver, Colorado, United States; hybrid with 2 days per week in the office, or fully remote anywhere in the United States outside the Denver market.
Base salary: $160,000–$180,000 per year, depending on experience and qualifications, plus equity.
Company
hirify.global operates a multi-tenant automotive SaaS platform serving thousands of dealer groups across the United States.
What you will do
Build and operate production agents on AWS Bedrock AgentCore using Strands or LangGraph.
Develop Python tools, MCP servers, and AgentCore Gateway targets for safe access to production APIs.
Create offline and online evaluation pipelines, LLM-as-judge CI gates, approval gates, and rollback paths.
Manage per-agent cost, latency, model routing, and production observability with AgentCore tracing, OpenTelemetry, and CloudWatch.
Own agent workflows from design through production and collaborate with Tech Leads on the domains and systems agents affect.
Evaluate business value and choose a deterministic tool or no agent when it is the better solution.
Requirements
5+ years of software engineering experience, including hands-on production experience building LLM agents in a code-first framework such as Strands, LangGraph, Google ADK, OpenAI Agents SDK, or a comparable framework.
Business-conscious judgment focused on customer value, including experience choosing a simpler solution over a technically clever one.
Strong Python skills and the ability to read and contribute to Java Spring Boot services.
Hands-on experience with AWS serverless services including Lambda, EventBridge, DynamoDB, and Step Functions.
Experience with prompt engineering, tool or function-calling design, and evaluating LLM output quality in production.
Readiness for a heads-down, hands-on building role measured by shipped code.
Nice to have
Direct experience with Strands Agents, AgentCore Runtime, managed agent runtime migrations, LLM-as-judge, agent-to-agent orchestration, or agent governance.
Production experience with MCP servers.
Experience owning OpenTelemetry and CloudWatch observability.
Automotive, fintech, or multi-tenant marketplace platform experience.
Familiarity with Bedrock Guardrails, LLM evaluation, model cost and latency optimization, multi-agent orchestration, Glue, Athena, ETL, or data lake tooling.
Culture & Benefits
Fun, casual, collaborative workplace with recognized contributions and direct business impact.
Denver hybrid with 2 days per week in the office or fully remote anywhere in the United States.
Employer-paid health, dental, vision, disability, and life insurance, plus family coverage starting on day one.
Unlimited PTO, 10 paid holidays, a 401(k) with company match, mental health support, and pet insurance options.
Equity grant, identity theft protection, and access to an on-site Denver gym and stocked kitchen.
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 →