Mlops Machine Learning Engineer Jobs in Miami, FL

You're interviewing for Opendoor’s ML team which seeks to automate and refine every decision made in our product. We don't slot into silos; you'll build where you have the most impact and the most fun.

These are builder roles across the ML stack. Wherever you land, you'll be doing one of three things:

• Building models in business-critical contexts like pricing, risk, repairs, and decision optimization. Leverage frontier techniques to extend our capabilities into the unstructured world of real estate.
• Building the intelligent services that bring structured, precise decision-making into the highly unstructured world of real estate.
• Building platforms that accelerate how fast our models learn. How fast we learn dictates how fast this company can grow.

You’ll work directly with researchers, product, and operations to build the automation that scales in the real world. Our systems must be agile, accurate, and resilient in a heterogeneous space. We are growing fast and this work is at the core.

This isn’t a role for everyone. We choose hard mode. We’re process-light, high-trust, and we don’t put artificial boundaries between you and the work. You’ll be expected to understand how your piece connects to the product and communicate at that level. We don’t have project managers, we don’t have scrum. We do reviews, proposals, demos, and trust.

You ship. You pick the boring solution when boring is right and the novel one when it isn’t. You know when “good enough and shipped today” beats “perfect next quarter.”

You have high agency. You don’t wait for permission or a perfectly scoped ticket. You see the problem, take ownership end-to-end, and pull in whoever you need. Lean teams, significant latitude, real accountability.

You run at unclear problems. The most valuable problems here don’t come with a playbook — messy data, imperfect ground truth, markets that shift under you. Ambiguity is the job, not an obstacle to it.

You hold a high standard. You measure twice and cut once. You review code, raise the bar on everything around you, and treat the quality of our end-to-end judgment as your problem.

You think in first principles. You have opinions on architecture, distributed systems, ML lifecycle tradeoffs, and the constraints and tripwires of operating models in a high-stakes environment.

You default to AI. You’ve already integrated modern AI tools into your daily workflow. You use them to move faster, not as a crutch.

You communicate well. You write clear design docs, give useful code reviews, push back on bad ideas without making it personal, and can land a technical tradeoff with a non-technical stakeholder.

You believe in what we’re building. Not hype, conviction. You see the opportunity in what we’re doing and you want to be part of finishing it.

You have fun. We stay human when times are hard. The task is daunting, but we’re all in it together.

• Build and train models that real customers and real money depend on — pricing, automation, and decision systems in production.
• Work side‑by‑side with researchers and analysts to turn prototypes into clean, testable, production‑ready code and systems.
• Own model pipelines end‑to‑end: data ingestion, training, validation, versioning, deployment, and monitoring.
• Design, build, and evolve mission‑critical services and APIs that connect to real‑world, messy operations.
• Build the platform that accelerates the full ML lifecycle: agentic research, automated retraining, experimentation, deployment, monitoring.
• Proactively tackle real‑world challenges like sparsity, data drift, and model decay in a volatile market.
• Use AI tools daily and help push them further than anyone else in the industry.
• Lead technical design reviews, mentor teammates, and raise the bar on everything around you.

• Senior‑level or above: deep experience shipping and operating production ML systems, ML‑adjacent services, or data/ML platforms.
• Strong fundamentals in Python; comfortable picking up new ones.
• Proficiency with statistics and ability to reason distributionally; has put it to work with real‑world monitoring of ML systems.
• Expertise with the end‑to‑end ML lifecycle (training, evaluation, deployment, monitoring, and iteration) and associated tooling (e.g. MLflow, Airflow, Spark, Delta Lake).
• Demonstrated ability to make and communicate design decisions and tradeoffs across stakeholders.
• Based in or willing to relocate to Miami, Toronto, or Seattle.

• ML systems experience in business‑critical domains: pricing, forecasting, logistics, marketplaces, risk.
• Streaming and event‑driven systems (e.g. Kafka), gRPC, Redis, or workflow engines.
• Interest in real estate or other messy, high‑stakes domains with imperfect data.

We move fast. Typically:

• A 60 minute technical deep dive to understand a past problem or project you’ve worked on.
• Two 60 minute pairing‑style technical reviews.

We’re not running these to see if you can finish a problem under pressure. We want to know what it’s like to work with you. Before each interview you’ll receive an email on what to expect.

Not a perfect fit on paper but clearly excellent? Apply anyway and tell us why in your cover letter. We value T‑shaped people. If you have deep expertise in an adjacent area and a strong point of view on how it applies here, that’s exactly who we want to talk to.