Lead AI Engineer

• Architect and lead development of AI-powered applications, especially those using agentic and LLM-based architectures
• Own and evolve end-to-end AI workflows, including RAG pipelines, prompt orchestration, tool usage, and safety layers
• Design and implement multi-agent systems with planning, delegation, memory, and inter-agent communication
• Oversee fine-tuning of foundational models for downstream tasks in various domains
• Define and monitor evaluation strategies to assess model performance, consistency, and reliability in production
• Lead implementation of prompt safety, content moderation, and guardrails across AI components
• Guide the integration of OCR and data extraction flows to process unstructured data
• Provide technical leadership across AI initiatives and mentor mid-level and junior engineers
• Collaborate with product and design teams to shape scalable, production-ready ML solutions
• Stay up to date with AI research and tools, balancing innovation with production readiness

• 5+ years of hands-on experience in AI/ML engineering, with at least 2 years building agentic or generative AI applications
• Proven experience leading complex AI initiatives in production, from architecture through deployment
• Deep expertise with LangChain, LangGraph, or OpenAI Agents and agentic workflows
• Strong understanding of foundation models, LLMs, and orchestration challenges
• Hands-on experience designing and deploying Retrieval-Augmented Generation (RAG) systems
• Experience with fine-tuning LLMs using supervised or reinforcement learning approaches
• Strong understanding of vector search systems (e.g., Pinecone, Weaviate, FAISS, Qdrant)
• Experience with evaluation pipelines and model monitoring in production environments
• Excellent communication skills and ability to lead cross-functional technical discussions
• Languages: Python (required), familiarity with JS/TypeScript is a plus
• LLM Tooling: LangChain, LangGraph, OpenAI Agents (required)
• Foundation Model APIs: OpenAI, Claude, HuggingFace, Cohere
• RAG systems: Context chunking, embeddings, metadata filtering, query optimization
• Vector databases: Pinecone, Weaviate, Qdrant, FAISS
• Model fine-tuning: LoRA, PEFT, prompt tuning, RLHF pipelines
• Evaluation frameworks: Prompt injection testing, regression evaluation, custom metrics
• Prompt safety & guardrails: Moderation layers, fallback chains, output validation
• OCR / Extraction: Experience with document parsing and unstructured data pipelines
• Leadership and mentoring mindset with technical authority
• Clear decision-making under ambiguity, balancing research with pragmatic delivery
• Systems thinker: can connect low-level tooling with product vision
• Strong ownership and stakeholder collaboration
• Fluent in English

familiarity with JS/TypeScript is a plus

• Flexible work schedule and remote work days
• ESOP (Employee Stock Ownership Plan) so that you grow with us
• Additional loyalty vacation days (up to 28 total vacation days based on experience and time with us)
• Health package via Regina Maria Clinic
• Meal tickets (40 RON/day)
• Discounted gym subscription
• Access to role-specific certifications/courses
• Regular team knowledge-sharing sessions and mentoring opportunities
• Clear visibility into your progress and opportunities for advancement
• Top-tier gear: Receive your own MacBook for high-performance development