Staff ML Platform Engineer (MLOps)
FutureFit AI · Remote · Remote (US) · Remote (Canada) ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Staff level position
- Fully remote (US/CA)
- Python + AWS SageMaker
- MLOps & LLM infra
- USD 172k-215k
You will own the end-to-end ML platform layer, ensuring models are built, deployed, evaluated, and served efficiently. This greenfield role requires establishing standards for compute provisioning, environment management, and cost-aware LLM routing while maintaining high system reliability.
What you'll do
- Evaluate current pipelines and create a prioritized plan for platform improvements
- Manage compute provisioning and keep training environments reproducible
- Build infrastructure for smart LLM routing to optimize cost and quality
- Implement disciplined A/B testing, shadow deploys, and canary releases
- Establish observability systems for drift detection and prediction traceability
- Debug production incidents and maintain model uptime alongside the team
- Ensure consistent feature computation and storage for training and inference
- Define deployment and monitoring standards for shared team ownership
What you'll need
- Staff-level hands-on experience in MLOps, ML platform, or infrastructure
- Proven track record setting up end-to-end MLOps practices including CI/CD
- Production experience with LLM systems, serving, and cost-latency tradeoffs
- Experience operating models in both batch and real-time serving contexts
- Proficiency with containers and managing reproducible environments at scale
- Ability to run controlled experiments like holdouts and define success criteria
- Deep knowledge of ML observability, lineage, and regression detection tools
- Comfort with data engineering tasks such as pipeline and feature store management
- Strong systems design skills to translate product needs into durable architecture
Nice to have
- Interest in crossing over into model development from platform engineering
- Experience shaping feature stores rather than just inheriting existing ones
- Background in evaluating AI observability vendors for buy versus build decisions
- Prior work in mission-driven, workforce, or government-adjacent data sectors
- Comfort mentoring small data and engineering teams during platform builds
Stack
- Python
- SQL
- AWS SageMaker
- Airflow
- dbt
- PostgreSQL
- Redshift
- MongoDB
- GitHub Actions
- Looker
How you'll work
- Fully remote for candidates based in the United States or Canada
- Optional office access in Toronto for local residents
- Travel expectation of two to three trips per year including off-sites
- Benchmarked compensation targeting the middle of the national market
What you get
- Competitive salary range aligned with institutional startup benchmarks
- Reasonable accommodations provided for individuals with disabilities
- Inclusive environment celebrating diversity across all protected characteristics
- Opportunity to shape core standards in a growing, mission-driven organization
Hiring process
- Online application submission
- Initial screen with Talent Acquisition team
- Interview with the Hiring Manager
- Performance challenge assessment
- Final one-on-one interviews
- Hiring decision typically reached within six weeks