Senior ML Platform Engineer | AI Training & CI/CD

ProducePay; United States; Hybrid; USD 184,000 - 272,000; Full time

Benefits offered by this job

Competitive compensation and equity

Health and Wellness benefits (Medical,

Unlimited Vacation

Flexible hours and Work from Home

Daily drinks, snacks and catered meals

Team building activities

Job summary

Waabi is seeking an experienced infra/MLOps engineer to expand our training stack and tooling for autonomous systems. You’ll optimize Kubernetes-based workflows, build developer-facing surfaces, and improve data pipelines across large multimodal datasets.

You’ll collaborate with researchers and platform teams to ensure reliable, scalable training at scale, implement observability, and ship CI/CD for models and simulations. Equity and competitive benefits apply.

Qualifications

• 5+ years of software or infrastructure engineering, including tools or platforms used by other engineers and operating ML or data-intensive production systems.
• Hands-on Kubernetes expertise — GPU scheduling, autoscaling, Helm or equivalent, networking fundamentals, and the ability to debug a cluster under load.
• Excellent Python, and a track record of designing APIs and CLIs other people enjoy using. AWS depth: object storage at scale, IAM, GPU compute, networking, and IaC (Terraform, Pulumi).
• Distributed training in PyTorch (DDP, FSDP, or similar) with experiment tracking and model registry tooling.
• Fluency with containers, CI/CD, and modern build systems, including large monorepos.
• The ability to influence without authority: evaluate frameworks, pilot credibly, and persuade engineers to change how they work.
• A collaborative default — you co-own a system rather than draw a boundary around your part.
• User empathy: fix the third-most-interesting problem blocking ten people.
• Strong product instincts, writing, and comfort operating autonomously in ambiguity.
• Passionate about self-driving technologies and frontier AI, and what a small team can do with the right infra.

Responsibilities

• Build and evolve our training infrastructure on Kubernetes with GPU scheduling, autoscaling, multi-node distributed jobs, and capacity strategy.
• Shape the developer-facing surface — CLIs, SDKs, job submission, templates, paved paths for teams.
• Shorten the inner loop: measure latency, iterate locally, and drive down time-to-signal.
• Evangelize best-in-class tooling with prototypes and migration paths.
• Strengthen the data and artifact layer: dataset versioning, sharding, high-throughput loading of large multimodal data.
• Turn one-off Python into durable tooling — libraries, CLIs, and observable services with sane defaults.
• Make experiments legible with dashboards, a model registry, and lineage from dataset to checkpoint.
• Ship CI/CD for models alongside autonomy and simulation.
• Build observability across the ML stack — utilization, throughput, and failure modes.
• Treat docs, onboarding, and support as product surface for researchers and engineers.
• Drive adoption with real users, iterate, and ship improvements quickly.
• Build guardrails with Security/IT/Infrastructure for IP-sensitive environments.

Skills

Kubernetes

Python

APIs & CLIs

PyTorch

AWS

CI/CD

Monorepos

Collaboration

Autonomous systems

Tools

Argo Workflows

Ray

Kubeflow

Slurm

Job description

Waabi is seeking an experienced infra/MLOps engineer to expand our training stack and tooling for autonomous systems. You’ll optimize Kubernetes-based workflows, build developer-facing surfaces, and improve data pipelines across large multimodal datasets.

You’ll collaborate with researchers and platform teams to ensure reliable, scalable training at scale, implement observability, and ship CI/CD for models and simulations. Equity and competitive benefits apply.