Research Engineer - Inference

📋 Description

• Deploy state-of-the-art models to production
• Own path from research checkpoint to serving infra
• Optimize inference latency, throughput, and cost
• Use quantization, distillation, KV-cache, batching, custom kernels
• Build high-performance serving systems for real-time workloads
• Create tooling to ship models quickly and safely

🎯 Requirements

• Experience deploying and serving ML models in production
• GPU programming and inference optimization: CUDA, Triton, TensorRT
• Experience with vLLM, SGLang or similar serving frameworks
• Ability to profile, diagnose, and eliminate bottlenecks

🎁 Benefits

• Innovative culture with impact-focused work
• Growth paths and opportunities to drive impact
• Learning & development stipend
• Social travel stipend for annual meetups
• Annual company offsite in new locations
• Monthly co-working stipend