AI Agentic Engineer

About the role

Our team designs and ships agentic systems that integrate into real business processes: agents that read documents, query internal systems, call external APIs, hand work off to one another, and know when to stop and ask a human.

Joining our team means working alongside knowledgeable researchers and engineers in an environment committed to staying at the forefront of agentic AI.

Requirements

• Strong Python skills, with at least 2 years of experience
• At least 3 delivered agentic projects
• At least 1 year of hands-on experience building and shipping LLM-powered agents, including tool calling, structured outputs, and multi-step reasoning
• Practical experience with an agent orchestration framework — LangGraph, OpenAI Agents SDK, Pydantic AI, or your own — plus LangChain and Langfuse or LangSmith in the surrounding stack
• Solid RAG fundamentals: chunking strategies, hybrid search, reranking, and the ability to identify when retrieval is the bottleneck
• An engineering approach to prompting: versioning, regression tests, and measurable evaluation rather than intuition alone
• Experience with observability and tracing for LLM applications, including debugging failed agent runs from traces and tracking token usage and latency
• Experience building REST APIs with FastAPI or a similar framework, including streaming responses
• Knowledge of Docker, relational databases, and vector databases
• At least upper-intermediate written and spoken English
• Ability to work from our Kyiv office at least 3 days per week

Responsibilities

• Translate open-ended business requirements into concrete agent architectures, choosing between a single agent, a multi-agent graph, and deterministic code
• Communicate directly with client-side managers and product owners
• Own agents end to end: design, implementation, evaluation, deployment, and iteration based on real-world use
• Build reliable tools and integration layers around agents, including APIs, databases, internal systems, and file pipelines
• Design and maintain evaluation harnesses and regression suites to measure the impact of changes to prompts, models, and graphs
• Harden agents against real-world failures with retries, fallbacks, guardrails, timeouts, graceful degradation, and human-in-the-loop checkpoints
• Optimize speed, cost, and flexibility through caching, model routing, appropriately sized models, and fewer round trips
• Write clean, reusable code and contribute to shared framework components that other teams build on
• Keep up with a rapidly changing stack and help the team distinguish useful developments from noise

Would be a plus

• Experience with voice-to-voice AI systems using LiveKit, Pipecat, or real-time model APIs, including latency budgets, interruption handling, and turn detection
• Experience with MCP (Model Context Protocol), including building servers, clients, or tool ecosystems
• Experience with computer-use or browser automation agents
• Experience evaluating deployed agents
• Fine-tuning experience with LoRA or QLoRA, or experience using small local models for routing, extraction, and classification
• Familiarity with emerging AI tools and marketplaces
• Knowledge of machine learning methods, including classical machine learning, recommendation systems, natural language processing, and computer vision
• Experience with edge ML

Working terms

• Startup culture, a goal-oriented team, and a research mindset
• The opportunity to apply your engineering skills to tools and systems for fellow engineers and help shape the future of AI
• Latest-generation MacBook Pro
• An in-house GPU cluster for training and experimentation
• 20 working days of annual leave
• English courses, educational events, and conferences
• Medical insurance

Tools & systems

Relevant experience matters more than knowing every tool listed.