About the roleWe are looking for a Founding ML Platform Engineer to design, build, and own the infrastructure that takes our machine learning models from research into reliable production systems.
You will establish the core MLOps platform at Parisi Labs, creating the tools and workflows that allow researchers to operationalize models without becoming responsible for the infrastructure around them. This includes model registries, dataset and experiment lineage, retraining workflows, scheduled batch inference, and production monitoring.
Our models operate on complex, time-dependent data, producing forecasts and simulations of physical systems. You will build the infrastructure that makes these workflows reproducible, observable, and dependable as our research and products evolve.
This is a hands-on founding engineering role with significant technical ownership. You will work directly with our CTO, Chief Scientist, and researchers to define the platform architecture, select technologies, and implement the systems the company will build on.
What you'll own- Design and build our end-to-end MLOps platform, from model registration and versioning through deployment, inference, and monitoring.
- Develop reliable, scalable batch inference infrastructure for recurring forecasts and model-generated simulations.
- Build model registries and lineage systems that track training inputs, dataset versions, model configurations, artifacts, and deployment history.
- Create standardized workflows and tools that allow researchers to register models, manage experiments, and move models into production.
- Develop automated retraining and deployment pipelines, including dataset preparation, validation, version promotion, and rollback.
- Build monitoring and observability for model performance, forecast accuracy, data drift, pipeline health, and production failures.
- Help steer the technical and research direction of the company.
What we're looking for- 6+ years of software engineering experience, including significant hands-on experience building and operating production ML systems.
- A track record of designing and owning ML infrastructure or MLOps platforms, ideally having built systems from the ground up.
- Strong Python and backend engineering skills, with experience building reliable data processing and distributed workflow systems.
- Experience with ML lifecycle tooling such as MLflow, orchestration frameworks such as Dagster, Airflow, or Prefect, and modern cloud infrastructure.
- Deep familiarity with model deployment, batch inference, model versioning, data lineage, and production monitoring.
- Comfort working closely with ML researchers and translating experimental workflows into reliable engineering systems.
- An entrepreneurial mindset, strong technical ownership, and the ability to operate independently in a fast-moving environment with limited structure.
Especially compelling backgrounds:- Experience as a founder, technical co-founder, or founding engineer at an early-stage startup.
- Experience building an ML platform from zero to production, rather than exclusively maintaining established infrastructure.
- Experience as an early technical hire who helped establish engineering architecture, tooling, and practices.
- Experience with time-series forecasting, energy systems, or other data-intensive scientific applications.
First 90 days- 30 days: Understand our research workflows, existing model infrastructure, and production forecasting systems. Identify the most important gaps in model lifecycle management, serving, and observability.
- 60 days: Establish foundational model registry and lineage capabilities, and deliver an initial standardized workflow for moving research models into production.
- 90 days: Own a reliable end-to-end production ML workflow, including model versioning, scheduled inference, monitoring, and clear operational visibility.
Location and compensationLocation: New York City. This is an in-person role.
Salary range: $200,000-$280,000 USD.
Equity: meaningful early-company equity.