Trainium (NKI) Kernel Expert (Remote) at Mercor

remote
Verified Opening · Sourced from official employer portal
Share with Batch & Groups: WhatsApp Telegram
Advertisement
Mercor
Remote · Posted 11 hours ago
Apply now

Referral disclosure: the Apply button uses TechJobs360’s Mercor referral link. If you are hired through it, TechJobs360 may receive a referral bonus from Mercor. You apply on Mercor’s own site, in the same way as applying there directly.

  • Pay: $70–$90 per hour (USD)
  • Type: Contract, hourly, remote
  • Open to: India
  • Hours: About 40 hours a week
  • Field: Software Engineering
  • Posted on Mercor: 2026-08-25

About the role

Evaluate the quality, correctness, and hardware-appropriateness of Neuron Kernel Interface (NKI) development tasks used to train and evaluate a frontier AI lab's models. You'll assess CUDA→NKI migration fidelity, Trainium-specific performance-optimization quality, and cross-platform numerical-correctness standards — and provide clear, rubric-based written feedback.

Basic Qualifications

  • 2+ years of hands-on experience developing or optimizing kernels using the Neuron Kernel Interface (NKI) targeting AWS Trainium/Inferentia2 hardware
  • Strong understanding of NKI-specific development patterns: tile-based computation, SBUF/PSUM/HBM memory-hierarchy management, partition-dimension constraints, and DMA orchestration
  • Demonstrated experience assessing CUDA→NKI migration quality
  • Familiarity with Trainium-specific performance profiling (NeuronCore pipeline utilization, tensor-engine throughput, memory-bandwidth bottlenecks)
  • Experience defining or evaluating cross-platform numerical-correctness standards (GPU vs Trainium accumulation order, rounding behavior, mixed-precision semantics)

Preferred Qualifications

  • Direct experience with AWS Neuron SDK, Neuron Compiler internals, or contributions to NKI kernel libraries
  • Prior CUDA or Triton kernel development
  • Familiarity with Trainium hardware specifications (NeuronCore-v2 architecture, on-chip SRAM topology, supported data types: FP32/BF16/FP8/INT8)
  • Experience benchmarking ML training workloads on Trn1/Trn2 instances

How applying works

You apply on Mercor’s site: create a Mercor profile, upload your CV and complete the steps the listing asks for, which usually include a short interview. Mercor decides who is selected, and pay is agreed with Mercor.

Apply on Mercor →

Browse more paid AI training roles in Worldwide Remote Jobs, or see all the platforms we list. Not the right fit? Alignerr, micro1, Ethos and Handshake AI also have paid remote AI work for professionals (referral links).

To apply for this job please visit work.mercor.com.

Get Notified for Similar Openings
Join 15,000+ tech engineers getting weekly verified off-campus & lateral hiring alerts.
Join Telegram
Trainium (NKI) Kernel Expert (Remote) at Mercor
Mercor · Remote (India eligible)
Apply Now →