Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
развернуть
$84к-162к/год
$110к$130кмедиана$170к
По похожим ролям (49) ~ $130к/год
На СНГ рынке: $1.1к/мес ($914 - $1.5к)
86
Хорошая вакансия
развернуть
У роли чёткий инженерный фокус, подробно описанные задачи и практическая работа с современной инфраструктурой AI-агентов; главное ограничение - работа доступна только из США.
Только США
Чёткие задачи
Современный AI-стек
Поддержка развития
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
AI Engineer I (Python/TypeScript/LLM): Building and maintaining a metric intelligence platform with an accent on deterministic Python skills, AI agents, backend services, and governed business metrics. Focus on evaluating agent quality, maintaining MCP connectors and data pipelines, and improving reliability through testing, monitoring, and CI/CD.
Location: Remote from anywhere in the United States for most positions; an in-person interview may be required during hiring
hirify.global is an insurance company focused on helping people reduce risk and recover from unexpected events, with more than 60 years of experience.
What you will do
Build, test, maintain, version, document, and retire Python skills and AI agents for the metric intelligence platform.
Create evaluation and regression checks, including test sets, validators, and output-format rules, and monitor agent usage, cost, failures, and quality.
Maintain the TypeScript-based Mastra agent runtime, including workflows, tools, memory, and model configuration.
Maintain MCP servers and connectors linking agents to the data warehouse, semantic layer, BI tools, and other systems.
Keep APIs, scheduled jobs, data pipelines, CI/CD, logging, and monitoring reliable; resolve failed jobs, connector issues, data-quality alerts, and user-reported problems.
Trace metrics to source data, make small reviewed changes to semantic models and the gold data layer, connect metric relationships, and write operational runbooks.
Requirements
Bachelor’s degree in a relevant quantitative, technical, engineering, or business-related field, or an equivalent combination of education and experience.
Typically 2 years of relevant experience in AI-native analytics.
Ability to independently write, test, and debug modular deterministic Python code.
Ability to read and modify an existing TypeScript/JavaScript backend codebase; TypeScript is required for the Mastra framework.
Working knowledge of SQL joins, aggregations, and window functions for warehouse analysis and metric validation.
At least one end-to-end working LLM application, such as a tool-calling agent, RAG application, or MCP server; a shipped side project counts.
Understanding of prompts, tools, retrieval, context limits, and nondeterministic agent output, with an evaluation-oriented approach to testing expected answers.
Daily Git and GitHub use, including branches, pull requests, and code review; ability to work with REST or JSON APIs and read API documentation independently.
Understanding of semantic layers, dimensions, metrics, dbt models, YAML configuration, and data lineage.
Nice to have
Master’s degree in a quantitative field.
Degree in computer science, engineering, mathematics, or statistics.
Experience implementing native AI solutions in the insurance industry.
Culture & Benefits
Remote flexibility from anywhere in the United States for most positions.
Paid vacation, sick time, nine company holidays, and volunteer hours.
Medical, dental, vision, life, and pet insurance, plus a 401(k) plan with company match.
Potential holiday, referral, and performance-based bonuses.
Education assistance, professional and personal development opportunities, promotional opportunities, and wellbeing resources.
Equal opportunity employment and reasonable accommodations for qualified applicants with disabilities.
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 →