Senior Machine Learning Engineer

Xenon7 is partnering with a top-tier life sciences client to deliver production-ready machine learning systems for research and manufacturing workflows. This contract Senior Machine Learning Engineer role is focused on building and operating end-to-end MLOps pipelines, shipping models through low-latency inference, and integrating ML into operational environments. The work is hybrid with 3 days onsite in the Indianapolis, IN area, supporting process ML integration, quality assurance, and facility automation.

What you’ll own

• Design, deploy, and maintain production-grade MLOps pipelines for continuous training, deployment, model versioning, and monitoring.
• Implement automated model drift detection and performance monitoring, including self-healing inference pipelines for high-reliability environments.
• Operationalize production ML models inside operational technology (OT), API manufacturing workflows, and chemical process control systems.
• Deploy predictive capabilities for batch processing, process control optimization, real-time quality assurance, and facility automation use cases.
• Build low-latency, high-throughput microservices and serving architectures using FastAPI, Triton Inference Server, and TorchServe.
• Containerize and orchestrate ML workloads across distributed cloud and edge systems with Kubernetes, Docker, and pipeline engines such as Kubeflow and MLflow.
• Collaborate with chemical engineers, computational biologists, and software architects to convert operational friction into production-ready ML solutions.
• Set enterprise-level MLOps standards, model governance, and CI/CD best practices across the full machine learning lifecycle.

What you bring

• Senior-level proficiency (10–20+ years) across software engineering, MLOps, production ML deployment, and infrastructure scaling.
• Proven experience deploying and maintaining production ML systems in non-standard, specialized domains, including transitions between process/chemical engineering ML and clinical/scientific research applications.
• Unrestricted US Work Authorization (no sponsorship available) and ability to work 3 days per week onsite in the Indianapolis, IN area.
• Pragmatic problem-solving, strong collaboration, and the ability to explain complex MLOps architecture to cross-functional engineering teams.
• Advanced Python and C++, plus deep proficiency with PyTorch, TensorFlow, or Scikit-learn.
• Hands-on expertise with Triton Inference Server, TorchServe, MLflow, Kubeflow, or Databricks ML runtime.
• Experience with Kubernetes, Docker, CI/CD pipelines, FastAPI/gRPC, and cloud ecosystems including AWS and/or Azure.
• Demonstrated ability building real-time model monitoring, feature stores, drift detection, and integration with enterprise data pipelines.
• Deep exposure applying ML in either scientific/clinical domains (drug discovery, small/large molecule, computational biology) or chemical/process engineering environments (API manufacturing, batch processing, SCADA/MES integration, process optimization).

Technologies you’ll work with

Python, C++, PyTorch, TensorFlow, Scikit-learn, Triton Inference Server, TorchServe, MLflow, Kubeflow, Databricks ML runtime, Kubernetes, Docker, CI/CD pipelines, FastAPI, gRPC, AWS, Azure, feature stores, SCADA, MES

Location and contract details

• Location: Indianapolis, IN Metro (Hybrid / 3-Day Onsite). Open to regional/EST candidates with onsite travel.
• Contract type: Contractor Full-Time / Enterprise Project Engagement (Outsourced via Xenon7).

Nice-to-haves

• Academic background in Chemical Engineering, Bio-process Engineering, Computer Science, or a related STEM discipline.
• Direct experience operationalizing ML models in regulated GxP environments in Life Sciences or Specialty Chemicals.
• AWS Certified Machine Learning – Specialty, Databricks Certified Machine Learning Professional, or equivalent MLOps credentials.

Role clarity: This is not an exploratory data science or Jupyter-focused position. It is a hands-on MLOps and software engineering lead role centered on production deployment, inference systems, and ML integration. It is also not fully remote, with a required hybrid commitment of 3 days onsite per week at the Indianapolis client site.