Reposted : Staff Software Engineer, Machine Learning

🧾 Position Summary

Title: Staff Software Engineer – Machine Learning

Location: Remote (U.S.-based only)

Base Salary: $272,000 – $306,000

Plus: Equity + Full Benefits

Team: Discovery/Engagement ML

📍 Mission: Build ML-driven systems to power recommendations, personalization, and discovery experiences that engage Discord's 200M+ monthly users.

🧠 About the Role

This is a senior IC (individual contributor) role focused on building state-of-the-art recommender systems. You’ll design and deploy low-latency, high-throughput ML models that impact discovery, notifications, and user engagement at massive scale.

Discord is looking for a technical leader who can shape architecture, solve systemic ML challenges, and drive product impact through intelligent personalization.

🎯 Key Responsibilities

• Architect and build scalable ML-powered systems (e.g., recsys, personalization, ranking).
• Use advanced techniques like:
• Transformer-based models
• Multi-task learning
• Deep reinforcement learning
• Optimize for both speed of iteration and long-term scalability.
• Lead complex cross-functional initiatives with Product, Data, and Engineering.
• Use real-world user behavior and feedback to shape modeling strategies.

✅ Must-Have Qualifications

• 8+ years in applied ML, including production deployment of large-scale models.
• Deep expertise in recommendation systems (two-tower, transformer-based, etc.).
• Mastery of Python and frameworks like PyTorch, JAX, TensorFlow.
• Proven record of:
• Designing end-to-end ML pipelines
• Scaling inference systems for millions of users
• Driving measurable product outcomes
• Strong communication and leadership in ambiguous or complex environments.
• Experience deploying ML in real-time, latency-sensitive applications.

🌟 Bonus Points For:

• Whole-page optimization or multi-objective ranking.
• Reinforcement learning in production (e.g., reward tuning, bandits).
• Building user embeddings, user modeling, or profile enrichment for personalization.
• Familiarity with distributed training frameworks (e.g., DDP, Ray, Horovod).
• Experience with large-scale streaming/data pipelines (Flink, Spark, etc.).

📌 Why This Role Is Important

Discord’s platform is growing far beyond just voice chat—it’s now a content-rich ecosystem of communities, creators, events, bots, games, and more. This role will help unlock the right experience for the right user at the right time, deeply shaping engagement, retention, and user satisfaction.

If you’ve worked on YouTube, TikTok, Spotify, Meta, Pinterest, or similar large-scale ML-driven platforms, your experience would be directly relevant here.

🧠 Should You Apply?

Absolutely yes, if:

• You’ve led production ML efforts for personalization, ranking, or recommendation
• You want to be a technical leader rather than move into management
• You care about performance, scale, and measurable product impact
• You want to apply cutting-edge ML research in a massive-scale environment