MLOps Engineer: Build & Scale AI Pipelines
Posted Updated
Description
Planitar, Inc. seeks an experienced MLOps Engineer to own the end-to-end platform that trains, evaluates, deploys and monitors floor-plan generation models. Join our Advanced Research Group to build scalable ML infrastructure and automate QA for AI-driven floor plans.
You will run experiment tracking, model registries, data versioning, and ship models into desktop drafting workflows with careful rollout and rollback. Hybrid work in Ontario supports collaboration with a focused research team.
Build an automated evaluation harness that scores every new model and gates releases on regressions. Automate and version the data generation and training pipelines. Set up experiment tracking and a model registry. Package and deploy models into our desktop drafting application and automated pipeline, with independent rollout and rollback. Operate LLM-based agents in production: prompt and model versioning, cost and latency tracking. Monitor model quality, drift and cost in production. Manage our GPU compute and CI/CD for ML. 4+ years in MLOps, ML platform or ML-focused software engineering, with models you took to production. Strong Python; able to work with C++ codebases. PyTorch training workflows and ONNX export. Experiment tracking, model registries and data versioning (e.g. MLflow, W&B, DVC). Docker, workflow orchestration and CI/CD. AWS and Linux GPU environments. Comfortable as the sole owner of a platform on a small research team. Flexible benefits Health and dental coverage Health Spending Account or Lifestyle 3% DPSP match Hybrid work environment Professional development incentive Growth opportunities REA Group backingSkills
Sign in to read and join the discussion.