Member of Technical Staff (RecSys)

About Astrocade

Astrocade is a UGC gaming platform where anyone can turn an idea into a playable, shareable game in days, not months. Think YouTube, but for games. We've grown to more than 20 million users within just a few months of launch, and we're only getting started. Backed by Sequoia, NVIDIA, and Google, Astrocade was founded by Amir Sadeghian (Stanford PhD), Ali Sadeghian (Ex-Google Research), and Fei-Fei Li (Godmother of AI). We're building the infrastructure for a new era of interactive entertainment.

About the Role

Our recommendation system doesn't exist yet, and it needs to. We have a large and fast-growing catalog of games, millions of users, and new content being created every day. What you show someone, and when, is the difference between a session that lasts two minutes and one that lasts two hours. As our RecSys founding member, you'll own this problem end-to-end - set the architecture, build the foundation, and grow it from rule-based systems to deep learning. The decisions made now will shape how discovery works on the platform for years.

You'll report to the CTO, and work directly with the co-founders. This is a 0 to 1 build with full ownership.

What You'll Do

• Design and build the systems that decide which games surface to which players, from candidate retrieval through final ranking

• Own the full data pipeline - ingestion, feature engineering, training data construction, and low-latency serving

• Build personalization systems and models that adapt to user behavior, preferences, and context over time

• Build eval infrastructure to measure recommendation quality: offline metrics, online experiments, and business outcomes

• Run A/B tests and translate results into concrete system improvements

• Instrument the recommendation stack deeply so the team can move fast with confidence

You'd Be a Great Fit If You:

• Have 4-7+ years of experience in recommendation systems, ML engineering, or applied ML in a consumer context

• Have built ranking or personalization systems end-to-end - feeds, video, gaming, or similar

• Experience running recommendation evals end-to-end (offline + online)

• Understand the full stack - data pipelines, feature stores, model training, and serving

• Are comfortable making architectural decisions on a greenfield system without much scaffolding

• Are self-directed and energized by ownership, not just execution

Bonus Points

• Experience at companies with large-scale consumer recommendation systems (YouTube, Netflix, TikTok, Instagram, LinkedIn, Twitter/X)

• Familiarity with both rule-based and deep learning approaches, and when to use each

• Background in UGC or creator platforms where content is high-volume and fast-changing

Compensation & Benefits

• Competitive base + equity + bonus

• Health, dental, and vision coverage

• Lunch provided daily

Join us to help build the future of interactive entertainment.