GenAI Engineer (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
GenAI Engineer (AI): Developing AI Agents to transform complex data into actionable intelligence with an accent on agentic AI integration, prompt engineering, and LLM architecture design. Focus on building evaluation frameworks for agentic AI and configuring specialized assistants for FP&A, S&OP, and Retail Planning.
Location: Hybrid in Madrid, Spain (3 days office / 2 days remote)
Company
provides financial and operational planning solutions powered by advanced analytics and AI-driven insights for global enterprises.
What you will do
- Define and implement the optimal agentic AI architecture for Assistants.
- Optimize prompts and configurations to ensure high-quality AI outputs.
- Design a framework for evaluating LLM mixtures and agentic AI architectures.
- Configure AI assistants tailored for flagship solutions such as FP&A, S&OP, and Retail Planning.
- Collaborate with product managers, UX, and engineering teams across the entire product lifecycle.
Requirements
- Solid experience in software architecture and engineering.
- In-depth knowledge of Generative AI (GenAI) and Agentic AI.
- Strong expertise in prompt engineering best practices and evaluation frameworks.
- Proven experience implementing, testing, and evaluating GenAI systems.
- System integration skills, particularly with LLM integrations.
- Familiarity with Agile development processes.
Nice to have
- Experience with Azure integrations.
Culture & Benefits
- Hybrid work schedule (3 days office, 2 days home) for balance between collaboration and productivity.
- Collaborative environment that values unique perspectives and a growth mindset.
- Strong commitment to diversity, inclusion, and belonging.
- Opportunity to work on cutting-edge AI technology within a customer-centric product team.
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 →