Founding ML Platform Engineer — Remote + Equity

Talantir; United States; Remote; USD 90,000 - 200,000; Full time

Job summary

Deep Reasoning Labs is building a deep reasoning layer for LLMs, focusing on scalable, reproducible training and evaluation. You will own the internal MLOps platform and the research inference layer, ensuring fast, auditable iterations from training to rollout.

This role emphasizes systems engineering, multi-GPU training, and cost-efficient GPU orchestration. Remote-first with US/Canada overlap and occasional in-person sprints in SF preferred.

Qualifications

• Strong systems + ML infra experience for end-to-end training/evaluation pipelines.
• Experience running LLM inference serving with batching and caching.
• Proficiency with multi-GPU training and cost-aware orchestration.

Responsibilities

• Own training + research MLOps platform and inference layer.
• Build cost-efficient, reproducible pipelines (SFT, verifier/PRM, RLVR).
• Manage data pipelines, dataset versioning, model registry, and experiment tracking.

Skills

Systems engineering

ML infra

PyTorch distributed

Cloud/IaC

Experiment tracking

Cost optimization

Profiling

Tools

Kubernetes

AWS/GCP

Terraform/CloudFormation

vLLM/SGLang

DeepSpeed

Job description

Deep Reasoning Labs is building a deep reasoning layer for LLMs, focusing on scalable, reproducible training and evaluation. You will own the internal MLOps platform and the research inference layer, ensuring fast, auditable iterations from training to rollout.

This role emphasizes systems engineering, multi-GPU training, and cost-efficient GPU orchestration. Remote-first with US/Canada overlap and occasional in-person sprints in SF preferred.