Conversational AI engineer for poker coaching

QuintAce is building an AI-powered poker learning experience. We are looking for a Senior Conversational AI Engineer to develop an intelligent poker coach that helps players understand decisions, review hands, and improve through natural conversation. This role combines hands-on AI engineering with strong poker expertise. You will translate poker strategy and analytical outputs into accurate, clear, and useful coaching conversations, working closely with product, design, and poker specialists.

• Build and improve conversational AI experiences for poker coaching, hand reviews, strategy explanations, and personalized learning.
• Develop LLM workflows that combine conversation context, player history, poker knowledge, and structured analytical tools.
• Connect the AI coach with solver outputs and other poker services, ensuring explanations accurately reflect the underlying data.
• Design reliable tool calling, retrieval, memory, and context management for multi-turn conversations.
• Build evaluation datasets and quality checks covering poker accuracy, reasoning consistency, coaching usefulness, and conversational quality.
• Reduce hallucinations and ensure the coach handles missing information, ambiguous hand histories, and uncertainty appropriately.
• Own features from prototyping through production, including monitoring, debugging, and improvements to latency, cost, and reliability.
• Collaborate in English with colleagues across engineering, product, design, and poker.

• Strong hands-on experience building and shipping production ML or AI systems.
• Practical experience developing LLM-powered conversational applications, including tool calling, retrieval, and multi-turn context management.
• Strong Python skills and experience integrating models, APIs, and backend services.
• Demonstrable poker expertise, particularly in Texas Hold’em: you can analyze hands and discuss ranges, position, pot odds, equity, expected value, bet sizing, and GTO versus exploitative adjustments.
• Experience using poker solvers or study tools, with the ability to interpret their outputs and explain the assumptions and limitations behind a recommendation.
• Ability to turn complex poker concepts into clear, actionable explanations suited to different player skill levels.
• Experience evaluating AI outputs systematically and using evidence to improve product quality.
• Strong spoken and written English: you can participate actively in technical discussions, write clear documentation, and assess the quality and accuracy of English-language poker coaching.
• Ownership, sound engineering judgment, and comfort working across disciplines.
• Будет плюсом: Experience building poker software, solvers, training tools, or hand-history analysis products; Experience with personalized learning, tutoring systems, or coaching products; Familiarity with fine-tuning, preference optimization, or reinforcement learning; Experience optimizing LLM inference, observability, and production serving; Knowledge of game theory and imperfect-information decision-making

• Условий в вакансии нет