ML Ops Engineer

About DockwareDockware is a logistics technology startup that creates a digital twin of every shipment, using hardware and advanced computer vision models to capture, validate, and update shipment data. We passively capture shipment data and drive the workflow integrations that separate good tech from good products. Dockware uses these models and inexpensive deployment formats to lower data acquisition costs for carriers and shippers, transform that data into a validated digital twin of every shipment, and dynamically update the shipment's information as it moves from origin to destination. Email jobs@dockware.ai with your resume if interested.About the RoleWe're looking for an ML Ops Engineer to build, deploy, and maintain the infrastructure that powers our core computer vision products. You will primarily focus on creating robust, scalable machine learning pipelines which include data versioning, model training/versioning, evaluation, and deployment, while supporting Vision (our autonomous sensor suite) and Scan (our mobile application).Vision App: Infrastructure supporting autonomous sensor suite's computer vision models.Scan App: Infrastructure supporting the mobile application's freight dimensioning and tracking.ML Pipelines: Design and own scalable pipelines for data versioning, labeling, model training, evaluation, and continuous deployment of custom computer vision models.Backend & infra: APIs, data pipelines, and deployment systems supporting all products, leveraging AWS and serverless compute.The role requires both building and testing new features as well as maintaining production systems. You'll ship infrastructure that directly supports computer vision models running on warehouse floors and in the hands of logistics operators daily.What We're Looking ForRequired:High ownership mentality and comfort with significant autonomy—you see problems through to resolution, whether that means debugging infrastructure or redesigning entire ML workflows.Expertise in building and scaling custom machine learning models in production environments.Strong experience with MLOps principles, including data and model versioning, labeling, training, and evaluation pipelines.Familiarity with at least some of our core stack and interest to learn the rest:MLOps Tools: MLflow, DVC, UltralyticsInfrastructure: Docker, AWS (Lambda, SQS, ECS), Serverless Compute, CI/CD, testing suites, Github Actions, K8s/K3s, Vertex AI, TerraformBackend: FastAPI, Pydantic, Postgres, SQLiteGenuine interest in solving hard problems in logistics and computer vision.Bonus Points for:Experience specifically deploying computer vision models (e.g., using technologies like ONNX or TensorRT).Track record of shipping production software systems in a high-autonomy environment.Experience with large-scale data pipeline construction and optimization.Experience in the shipping or logistics industry.What Makes You Successful HereYou don't get stuck in the weeds. You commit and follow through. You're comfortable with ambiguity and can make progress without complete specifications. You ask "why" before diving into "how," and you're not afraid to challenge assumptions when something doesn't make sense.This is a high-ownership environment where your work directly impacts customers and revenue. You'll have autonomy, but you'll also be expected to take initiative and drive projects to completion.LocationStriking distance of Tulsa, OK preferred, but we're open to exceptional candidates regardless of location. This role may require periodic travel to customer sites and our Tulsa headquarters.Dockware is building the primary data capture layer for global logistics. If you're ready to tackle computer vision challenges at warehouse scale while keeping world-class engineers aligned and shipping, we want to hear from you.