AI Product Manager

Location: Hybrid in Seoul, South Korea

Company

hirify.global is building a proactive AI assistant that helps users manage conversations, errands, organization, and everyday workflows.

What you will do

• Define end-to-end AI system requirements, from model capabilities and system behavior to user impact.
• Translate model capabilities, data constraints, and evaluation results into product and system decisions.
• Make trade-offs across quality, latency, cost, reliability, and user experience.
• Collaborate with ML, backend, and mobile engineers on system design, evaluation, and iteration.
• Define evaluation frameworks spanning offline metrics, online experiments, and human feedback.
• Own product quality, execution, prioritization, correctness, predictability, and user trust.

Requirements

• Strong computer science fundamentals, including algorithms, data structures, and system design.
• Solid understanding of machine learning fundamentals and modern AI systems in production.
• Hands-on experience with AI-powered products, including LLM-based systems.
• Experience with model evaluation, prompt or pipeline iteration, and feedback loops.
• Significant experience owning complex technical products end to end and working closely with senior engineers and ML teams.
• Strong judgment in ambiguous, fast-moving environments and the ability to balance ambition with technical and operational reality.

Nice to have

• Experience shipping AI-heavy consumer products.
• Background as an engineer or highly technical product manager.
• Experience defining evaluation metrics for ML systems.
• Experience with AI UX patterns, failure handling, and zero-to-one product environments.

Culture & Benefits

• High-talent-density, hands-on team with a focus on practical AI products.
• Collective decision-making and rapid execution.
• Emphasis on structure, independent execution, quality, and continuous learning through user feedback.
• Interviews are conducted via virtual meetings and/or onsite.

Hiring process

• Three, and no more than four, interviews may be scheduled if there is a potential fit.
• Applications are evaluated by technical team members.
• Decisions are expected to be made promptly after the interview process.