Staff Engineer, Model Training Infra & Platform (Remote)

micro1; United States; Remote; USD 200,000 - 320,000; Full time

Benefits offered by this job

Equity

Health premium reimbursement

401K match

Remote-first

Job summary

micro1 is seeking a Member of Technical Staff, Model Training (infra & platform) to own distributed training platforms and GPU infrastructure for researchers. You will drive architecture decisions across training, inference, and simulation environments, enabling faster experiments and high-performance model deployment.

The role emphasizes technical ownership, cross-functional collaboration, and mentoring while delivering robust, scalable infrastructure for ML workloads in a remote-first setting.

Qualifications

• Strong CS fundamentals in distributed systems, concurrency, networking, and performance engineering.
• Experience designing and operating scalable backend systems or distributed computing platforms.
• Proficiency in Python and a systems language (C++, Rust, or Go).
• Hands-on experience with GPU-based ML infrastructure, distributed training, or large-scale compute orchestration.
• Strong understanding of system performance, memory efficiency, and fault tolerance.
• Experience designing reliable APIs, services, and infra components for complex workflows.
• Demonstrated technical ownership and cross-functional leadership across teams.
• Ability to thrive in fast-moving research environments with evolving requirements.

Responsibilities

• Architect, build, and scale shared infrastructure for supervised fine-tuning (SFT), RL, and on-policy distillation for language, multimodal, and robotics workloads.
• Design distributed training and execution pipelines coordinating GPU workers, rollout generation, teacher-model inference, reward computation, and simulations.
• Develop high-performance infra for model training and inference, optimizing GPU utilization and memory management.
• Build and evolve core architecture of micro1's agent ecosystem, including runtimes, interfaces, and APIs.
• Develop robust systems for experiment tracking, versioning, checkpointing, fault recovery, and observability.
• Identify and eliminate bottlenecks to reduce experiment turnaround time for complex workloads.
• Define standards, interfaces, and architecture patterns to support collaboration across teams.
• Lead engineering initiatives from design to production, mentoring engineers while owning challenging components.

Skills

Distributed systems

Concurrency

Python

C++/Rust/Go

GPU ML infra

API design

Tools

PyTorch Distributed

DeepSpeed

Ray

Kubernetes

NCCL

CUDA

Job description

micro1 is seeking a Member of Technical Staff, Model Training (infra & platform) to own distributed training platforms and GPU infrastructure for researchers. You will drive architecture decisions across training, inference, and simulation environments, enabling faster experiments and high-performance model deployment.

The role emphasizes technical ownership, cross-functional collaboration, and mentoring while delivering robust, scalable infrastructure for ML workloads in a remote-first setting.