Inference Engineer (AI)

Location: Remote or in-person at offices in San Francisco, United States

Salary: $150,000–$350,000 annual cash compensation range, including equity incentives

Company

hirify.global is building an open superintelligence stack combining compute, environments, evaluations, secure sandboxes, training, and deployment for frontier AI research.

What you will do

• Extend the prime-rl and vLLM integration to support new model architectures in reinforcement-learning training.
• Improve rollout throughput for long-horizon agents by tuning concurrency and KV-cache reuse.
• Optimize multi-node mixture-of-experts inference, expert parallelism, and prefill/decode separation.
• Reduce trainer-to-inference weight-update overhead through sharding-aware transfers and communication overlap.
• Build on FP8 and evaluate lower-precision inference while detecting numerical regressions.
• Collaborate with model researchers on architecture choices that balance model quality and hardware execution cost.

Requirements

• Experience building or optimizing GPU inference systems and understanding inference-engine internals.
• Strong Python skills and experience with PyTorch or a comparable machine-learning framework.
• Ability to profile end-to-end workloads and verify performance improvements through measurement.
• Experience debugging numerical or integration issues in machine-learning systems.

Nice to have

• Contributions to vLLM or another inference engine.
• CUDA or Triton kernel development, particularly for low-precision inference.

Culture & Benefits

• Flexible work arrangements with remote or in-person work options.
• Visa sponsorship and relocation assistance are available for international candidates.
• Quarterly team off-sites, hackathons, conferences, and learning opportunities.
• Opportunity to work on frontier AI infrastructure with a mission-driven research and engineering team.