Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указана
На похожих ролях по рынку
$190к$225кмедиана$260к
49 в выборке
На СНГ рынке: $4.2к/мес ($3.1к - $5.1к)
71
Хорошая вакансия
развернуть
Вакансия предлагает работу с production-системами на современном AI-стеке, но зарплата не раскрыта, а в консалтинге проекты и работа с клиентами могут меняться.
Кликните для подробной информации
Зарплата не указанаРабота с клиентамиСовременный AI-стекФокус на ИИВиза и страховка
Оценка от Hirify AI
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Applied AI Engineer – Agentic Systems (AI): Design, build, and deploy LLM-powered autonomous agents and multi-step workflows for enterprise use cases with an accent on tool calling, state management, context engineering, and secure integrations. Focus on building production-grade Python systems, evaluating agent performance, controlling hallucinations and latency, and translating complex business processes into reliable AI architectures.
Location: On-site in Dubai, United Arab Emirates
Company
hirify.global is a UAE-based AI and data consultancy delivering secure, scalable intelligent systems for enterprise and financial services clients.
What you will do
Design and implement multi-step agent loops, state machines, and tool-calling mechanisms for autonomous enterprise workflows.
Build context-management pipelines and RAG modules grounded in dynamic enterprise knowledge sources.
Develop secure integrations between LLMs, agents, enterprise databases, legacy business applications, and modern protocols such as MCP.
Write production-ready Python with automated testing, structured output validation, fallback logic, and human-in-the-loop controls.
Instrument tracing and evaluation suites to benchmark performance, diagnose hallucination loops and latency spikes, and optimize token costs.
Collaborate with client technical teams and business sponsors to define technical designs, acceptance criteria, and system architectures.
Requirements
Demonstrable personal experience implementing AI agents, multi-agent frameworks, tool calling, and state-management systems.
Advanced Python proficiency focused on maintainable code, pytest testing, code reviews, and Git version control.
Practical experience integrating commercial and open-weight LLM APIs, designing prompts, validating structured outputs, and handling model and tool failures.
Experience evaluating LLM applications with representative task sets, failure-mode analysis, regression testing, and application-tracing tools.
Ability to balance reasoning quality, system reliability, inference latency, and operational cost.
Nice to have
Experience deploying and supporting LLM agents used by active business teams or external customers.
Familiarity with LangGraph, AutoGen, CrewAI, LlamaIndex Workflows, or custom state-machine implementations.
Hands-on experience with MCP, Docker, and cloud deployment on AWS or Azure.
Prior experience delivering technical solutions in regulated enterprise environments such as banking, financial services, or insurance.
Culture & Benefits
On-site work in Dubai within a specialist AI and data consultancy.
Gold Level personal health insurance and individual visa sponsorship.
Professional development, certification support, and role-related subscription reimbursement.
Opportunity to work on cutting-edge AI projects with career advancement opportunities.
Monthly employee incentive program and an inclusive environment supporting diverse thinking styles.
Hiring process
Be prepared to walk through a system personally built, including implementation choices, evaluation methods, system limitations, and the path from prototype to production.
The application and interview process is designed to be accessible, predictable, and fair, with reasonable adjustments available upon request.
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 →