Lead Machine Learning Engineer – Recommendation Systems

GRAI is a social music app out of Warsaw, currently in alpha. We’re building new ways to listen and respond through music - focusing on the social, human interactions that only happen when people are connected on both ends. To make those connections truly resonant, we’re looking for a Lead Machine Learning Engineer to architect, build, and scale our recommendation and discovery engines. You’ll own the technical roadmap for our RecSys stack, transforming raw user behavior and social signals into deeply personalized, real-time experiences. WHAT YOU’LL DO - Technical Leadership: Define the long-term technical vision and architectural roadmap for our recommendation and personalization engines. - System Architecture: Design and oversee the implementation of scalable retrieval, ranking, and re-ranking pipelines capable of handling massive user behavior and content data. - End-to-End Ownership: Lead the development of robust ML infrastructure, including automated data processing, feature stores, MLOps), and real-time model monitoring. - Mentorship & Culture: Mentor and coach a talented team of ML engineers, fostering a culture of technical excellence, continuous learning, and rigorous experimentation. - Data-Driven Strategy: Design comprehensive offline evaluation frameworks and lead complex A/B testing strategies to validate and iterate on model performance. - Cross-Functional Collaboration: Partner closely with Product and Engineering to align ML initiatives with high-level business metrics and product goals. WHAT WE’RE LOOKING FOR - Proven Track Record: Extensive experience designing, building, and scaling production-grade recommendation systems, search engines, or large-scale ranking models. - Technical Mastery: Deep, foundational knowledge of machine learning, deep learning architectures, and modern information retrieval methodologies. - Scale & Infrastructure: Significant experience with distributed data systems and modern ML frameworks (PyTorch, TensorFlow). Proven ability to handle massive, high-throughput user interaction datasets. - Production & MLOps: Strong background in deploying and maintaining low-latency models in production, with a solid grasp of feature stores, model registries, and drift monitoring. - Leadership Skills: Demonstrated experience leading technical projects, mentoring engineers, and successfully managing stakeholders without losing your hands-on technical edge. - Pragmatic Execution: Ability to balance cutting-edge AI experimentation with the practical realities of production stability, latency constraints, and business value. NICE TO HAVE - Real-Time Expertise: Experience with streaming data architectures and real-time/session-based recommendation systems. - Domain Knowledge: Background in audio processing, music streaming, or high-growth consumer-facing personalization products. - Advanced ML: Familiarity with graph neural networks (GNNs), reinforcement learning, or leveraging Large Language Models (LLMs) for recommendation context. WHY JOIN US - High Autonomy & Impact: Direct ownership over the technical direction of high-impact ML systems used by millions of users. - Founding-Stage Equity: You’re joining a tight team at the alpha stage. We offer meaningful stock options so you have real skin in the game and share directly in the upside of what we build. - Influence the Future: Shape not just the recommendation stack, but the broader engineering culture and hiring roadmap of a fast-growing startup. - Collaborative Innovation: Work in a supportive, product-driven environment where your ideas directly dictate the future of how people experience music. - Rest & Recharge: High-output work requires real downtime. We offer 26 business days of paid time off per year, plus 5 days off and Polish public holidays.