AI Developer with Python for Customer Care AI Platform team (hybrid)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
AI Developer (Python): Building a next-generation multimodal AI platform for customer care with an accent on agentic workflows, RAG/CAG pipelines, speech-to-speech interactions, and hybrid LLM integration. Focus on designing low-latency real-time systems, optimizing context retrieval, and developing automated QA for continuous improvement and alignment.
Location: Bucharest (hybrid)
Company
Leading European provider of cloud infrastructure, hosting services, and AI-powered customer support tools serving millions of customers across multiple markets.
What you will do
- Design and implement LLM-based agentic workflows with tool usage, orchestration, and secure interactions via Model Context Protocol (MCP).
- Build and optimize Retrieval-Augmented Generation (RAG) and Context-Augmented Generation (CAG) pipelines for accurate, low-latency responses.
- Develop real-time Speech-to-Speech (S2S) pipelines focusing on streaming, latency reduction, and natural conversational flow.
- Create automated QA and evaluation modules using LLM-as-a-judge to measure and improve accuracy, safety, and resolution quality.
- Integrate and manage commercial and open-source LLMs, selecting models based on performance, cost, and use-case needs.
Requirements
- Location: Bucharest (hybrid work)
- Advanced Python skills including async programming (FastAPI, asyncio) and high-performance services; streaming architectures advantage.
- Experience building LLM systems with multi-agent orchestration (LangChain, LlamaIndex, LangGraph).
- Deep knowledge of vector databases (Weaviate, Qdrant, pgvector), semantic search, RAG pipelines.
- Real-time low-latency systems design, event-driven architectures, performance optimization.
- LLM evaluation frameworks, automated QA, LLM-as-a-judge patterns; RESTful API design, OAuth2.
Nice to have
- Java engineering skills.
- Experience with commercial models (OpenAI, Anthropic, Google) and open-source (Qwen, DeepSeek).
Culture & Benefits
- Flexible work schedule and hybrid option.
- 25 vacation days, substitute days for weekend holidays, meal tickets.
- Medical services package, access to trainings and e-learning platforms.
- Team events, networking, Summerfest/Winterfest parties, office perks (coffee, fruits).
- Open structures, flat hierarchies, strong team spirit, product responsibility.
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 →