Lead Machine Learning Engineer, Recommendation Systems

📋 Description

• • Own the end-to-end architecture and delivery of the next-generation recommendation engine that powers content discovery for 30 million monthly visitors across Launch2 Inc.’s portfolio of digital media brands.
• • Lead a cross-functional squad of 5-8 ML engineers, data scientists, and backend developers; set technical direction, define sprint goals, and ensure every release improves click-through-rate, dwell time, and lifetime value.
• • Architect real-time, low-latency inference pipelines that serve 50,000+ requests per second with <100 ms p99 latency while keeping cloud costs flat year-over-year.
• • Design and productionize hybrid models that blend collaborative filtering, content-based signals, graph neural networks, and contextual bandits to surface the right article, video, or newsletter at the right moment.
• • Establish rigorous offline experimentation frameworks (A/B, interleaving, counterfactual evaluation) and champion a culture of measurement where every feature is shipped behind a flag with pre-defined success criteria.
• • Partner with Product, Editorial, and Ad-Ops to translate business objectives into machine-learning objectives—e.g., balancing subscription conversions with ad-impression yield—while maintaining reader trust and brand safety.
• • Build self-service feature stores and model registries that allow data scientists to iterate on new embeddings without touching production code, cutting experiment cycle time from weeks to hours.
• • Drive data-quality initiatives across 500 TB of clickstream, content metadata, and user-profile data; implement anomaly detection and freshness SLAs that keep training sets reliable and bias-free.
• • Mentor senior and mid-level engineers through weekly 1:1s, architecture reviews, and brown-bag sessions; create clear progression rubrics that tie promotions to measurable technical impact.
• • Present quarterly roadmaps and results to the executive team, translating complex model performance into revenue, retention, and margin impact that informs company-level OKRs.
• • Champion privacy-first ML: implement differential privacy, on-device inference, and consent-aware feature engineering to stay ahead of GDPR/CCPA and browser cookie deprecation.
• • Contribute to the external tech community via open-source projects, conference talks, and blog posts that position Launch2 Inc. as a thought leader in responsible recommendation systems.
• • Continuously scan the horizon for emerging techniques—large-scale sequence models, reinforcement learning for slate optimization, synthetic data generation—and run quick spikes to validate ROI before full investment.
• • Collaborate with DevOps and FinOps to right-size GPU clusters, adopt spot-instance training, and build auto-scaling policies that cut infra spend by 20% without sacrificing model freshness.
• • Establish disaster-recovery and rollback playbooks that guarantee 99.9% uptime even during major model updates or traffic spikes driven by viral content events.