Architect and implement an AI-based productivity analytics system, covering data pipeline design through to final analytical outputs that support data-driven team decisions.
Construct scalable pipelines for large language models like Claude and GPT, including data segmentation strategies, parallel processing using MapReduce methods, and generating structured insights.
Develop a meta-workflow framework where language models autonomously generate, validate, and deploy automation scripts in isolated environments, featuring self-correcting mechanisms and zero manual deployment.
Design and maintain a system-level prompt engineering infrastructure, including a reusable library of prompt templates for diverse analytical use cases such as summaries, profiles, and performance deep dives.
Build a quality assurance framework for AI outputs, including hallucination detection, consistency evaluation, and regression testing to ensure reliable and repeatable results.
Extend the platform’s capabilities to new domains and analysis types without proportional increases in manual work, using configuration-driven modular architecture.
Document AI system design, define automation requirements, and communicate technical insights to stakeholders and clients.

Minimum of 2 years of production experience with large language models, including prompt engineering, pipeline development, and API integrations, with at least 1 year involving advanced features such as tool calling, structured outputs, and agent systems.
Proficient in Python, including async programming, dataclasses, type hints, and API integrations, with a focus on writing clean, testable, and maintainable code.
Familiarity with MapReduce patterns in LLM processing, including selecting data chunking strategies, managing parallel execution, and synthesizing results into unified analytical reports.
Hands-on experience developing agentic systems involving tool usage, self-correction loops, and multi-agent workflows, with the ability to assess when agents outperform fixed pipelines.
Strong SQL skills and experience working with analytical databases.
C1-level English proficiency, sufficient for reading technical documentation, writing specifications, and communicating asynchronously.
Proactive ownership mindset: capable of independently driving tasks from start to finish, debugging live issues, and iterating without close supervision.

Experience with workflow orchestration or automation platforms such as Windmill, Dagster, or Prefect, with knowledge of building robust automated systems.
Knowledge of retrieval-augmented generation (RAG) systems, vector databases, and embedding pipelines, with the ability to process large volumes of unstructured data.
Background in developing evaluation frameworks for LLMs using tools like LangSmith, PromptFoo, or custom solutions to measure and enhance AI output quality.
Familiarity with modern data platforms such as Databricks, Delta Lake, or Snowflake.
Experience in fast-paced, product-focused technology companies or AI startups requiring rapid development and iteration.
Understanding of product metrics including daily active users, retention, and unit economics, with the ability to align technical work with product outcomes.

Contribute to building Ukrainian-developed products that compete globally.
Freedom to implement original ideas and drive innovation beyond routine tasks.
Work on products designed for meaningful impact rather than financial returns alone.
Collaborate with a team of motivated, results-oriented professionals.
Engage with a global market and serve a large user base.
Access partner programs and best practices from major tech companies including Snapchat, Meta, Google, and Apple.
Established internal processes that support career growth without artificial barriers.
Performance reviews based on transparent, individual KPIs.
Competitive pay and a full benefits package.

Competitive compensation and a comprehensive benefits package

Remote (Worldwide) — Kyiv, London, Limassol, Warsaw

A team of like-minded, results-driven people working on impactful products with global reach

C1-level English proficiency — able to read documentation, write technical specifications, and communicate effectively in asynchronous environments.

AF is hiring an AI Engineer