aipythonsnowflakekubernetesazurellmragci cdobservabilityagentic ai
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$64к$111кмедиана$139к
19 в выборке
На СНГ рынке: $4.3к/мес ($3.8к - $5.6к)
88
Крутая вакансия
развернуть
Здесь можно отвечать за хорошо описанную корпоративную AI-платформу и работать с промышленными GenAI-решениями, облаком, данными и управлением рисками; главный компромисс - минимум три дня в неделю в офисе.
?Много направлений
Передовые AI-технологии
Влияние на бизнес
Офис три дня
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
AI Engineer (Agentic AI/Azure): Building Hero’s enterprise AI foundation, including an agentic AI platform, governed MCP Gateway, Azure knowledge base, and observability layer with an accent on production-grade GenAI applications, secure enterprise integrations, and scalable retrieval. Focus on designing agent orchestration, RAG pipelines, authorization and audit capabilities, and evaluation and cost monitoring for AI solutions governed by EU AI Act, GDPR, nFADP, and ISO 42001 requirements.
Location: Lenzburg, Switzerland; hybrid with a minimum of 3 days onsite
Company
hirify.global is an international family-owned company with Swiss roots that develops consumer food products and is building a group-wide AI Centre of Competence.
What you will do
Design, build, and evolve an enterprise Agentic AI Platform covering agent orchestration, tool use, RAG, prompt and agent management, model selection, CI/CD, and LLMOps.
Build the MCP Gateway as a governed access layer with a tool registry, authorization model, execution proxy, audit stream, and integrations with SAP S/4HANA, Microsoft 365, and SharePoint.
Develop an Azure-based AI Knowledge Base with ingestion pipelines, chunking, embeddings, hybrid retrieval, permission-aware filtering, Snowflake integration, semantic layers, and enterprise ontology.
Establish observability for agent and model quality, latency, failures, costs, usage, guardrail events, evaluation, tracing, logging, and auditability.
Develop proofs of concept and MVPs with the AI Use Case Manager and business owners, reusing platform capabilities and taking solutions into production.
Requirements
Degree in computer science, software engineering, applied mathematics, or a related field.
At least five years of software engineering experience, including at least two years building GenAI or LLM applications operating in production.
Hands-on experience with agentic frameworks and the Model Context Protocol.
Understanding of reusable platforms combining agent orchestration, RAG, tool use, model selection, evaluation, and prompt engineering.
Strong Python skills; practical experience with Microsoft Azure technologies including AI Foundry, AI Search, API Management, ADLS Gen2, and Entra ID.
Experience with Snowflake, SQL, and dbt, plus understanding of semantic layers and knowledge graphs.
Experience packaging and operating production-grade solutions with Docker, Kubernetes, infrastructure as code, and CI/CD through GitHub Actions or Azure DevOps.
Experience monitoring production systems and understanding how evaluation, tracing, logging, cost metering, and auditability support reliable AI.
Nice to have
Experience with TypeScript or C#.
Experience with AWS Bedrock.
Culture & Benefits
Freedom, substantial responsibility, flexible working hours, and a flat hierarchy.
40% home office, social benefits, and a modern working environment.
Exposure to senior stakeholders, IT leaders, data experts, and business teams at global scale.
Free snacks and drinks, subsidized meals, and free parking.
Applications are accepted only through the designated channels; applications from third-party job offer scrapers are not accepted.
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 →