Senior Machine Learning Engineer

Data; Northern (KY); Hybrid; USD 140,000 - 190,000; Full time

Job summary

Candid is seeking a Senior ML Engineer to take operational ownership of deployed ML/AI services, ensuring reliable production performance and governance. You will partner with data scientists to translate research into scalable deployments, implement CI/CD for ML, and optimize AWS-based infrastructure for cost and reliability.

You will build observability dashboards, manage model versioning and artifact handling, and work closely with product and software engineering teams to define integration

Qualifications

• 4+ years of professional software engineering, with at least 2 years in production ML systems (MLOps or production ML focus).
• Strong Python production-quality service code.
• Hands-on experience with experiment tracking and model lifecycle tooling in production (MLflow, Weights & Biases).
• Experience deploying PyTorch models in production with optimization techniques (quantization, batching, ONNX Runtime).
• Experience deploying and monitoring ML models in production, understanding drift, retraining triggers, and artifact management.
• Familiarity with AWS ML deployment stack (Lambda, ECS/Fargate, S3, IAM, CloudWatch).
• Experience building or operating CI/CD pipelines for ML services.
• Proven track record of improving production reliability via observability and disciplined deployment.
• Ability to collaborate with data scientists and translate their outputs to production decisions.
• Strong written and verbal communication; comfortable working independently.

Responsibilities

• Take operational ownership of deployed ML/AI services, monitor for issues, retraining cadences, and handoffs from data scientists.
• Improve inference performance of deployed models, including complex graph inference, via quantization, batching, and efficient serialization.
• Design and operate experiment tracking, model versioning, and artifact management for low-friction handoffs.
• Build and maintain observability: logs, metrics, dashboards, alerts to reduce incidents through good deployment hygiene.
• Establish repeatable deployment paths for new ML services on AWS with CI/CD and infrastructure-as-code patterns.
• Monitor AWS spend for ML workloads and publish cost visibility to the team (Bedrock, S3 lifecycle, etc.).
• Collaborate with data scientists to translate research code into production-ready deployments and surface operational insights.
• Serve as liaison between Data Science and product/software engineering teams to define integration contracts and SLAs.

Skills

Python

MLOps

PyTorch

AWS

CI/CD

Observability

Communication

Cross-functional

Model lifecycle

MLflow

Tools

MLflow

Weights & Biases

ONNX Runtime

Lambda

ECS/Fargate

S3

IAM

CloudWatch

Job description

Candid’s Data Science and AI team builds and ships machine learning and AI models across many concurrent projects. The Senior ML Engineer exists to take operational ownership of Candid’s deployed ML and AI services to free up data science capacity.

Responsibilities

• Take operational ownership of Candid’s deployed machine learning and AI services — monitoring for degradation, managing retraining cadences, coordinating handoffs from data scientists, and serving as the accountable point of contact for models in production.
• Improve inference performance for deployed models, including complex graph inference models, applying techniques such as quantization, artifact slimming, batching, and efficient serialization.
• Design and operate experiment tracking, model versioning, and artifact management so data scientists have a consistent, low-friction way to hand off work.
• Build and maintain observability for ML and AI services — centralized logging, metrics, dashboards, and alerting for the services that matter most — and proactively reduce incidents through good deployment hygiene.
• Establish and iteratively improve a repeatable deployment path for new ML services on AWS, including CI/CD integration and infrastructure-as-code patterns.
• Monitor and actively manage AWS spend across ML workloads, including Bedrock token usage, compute sizing, S3 lifecycle, and publish cost visibility to the team.
• Work closely with data scientists to understand model behavior, surface operational insights back to them, and translate research code into production-ready deployments.
• Serve as the primary technical liaison between the Data Science team and Candid’s product and software engineering teams when ML or AI services are integrated into products and systems, defining integration contracts, APIs, latency and reliability expectations, and input/output schemas, and supporting those teams through integration.
• Build and operate Amazon Bedrock-backed services and integrations where they are the right tool for the job.
• Contribute to Candid’s secure-by-default posture for ML services, including IAM scoping, secrets handling, and compliance-aligned tagging.
• Participate in technical planning and roadmap discussions for the Data Science team.

Skillset

• 4+ years of professional software engineering experience, including at least 2 years in a role where production ML systems were a primary responsibility (MLOps, ML engineering, or production-ML-focused data science).
• Strong proficiency in Python, including writing production-quality service code.
• Hands-on, operational experience with experiment tracking and model lifecycle tooling in production, such as MLflow, Weights & Biases, or comparable systems.
• Hands-on experience deploying PyTorch models in production, including familiarity with serving optimization techniques such as quantization, batching, ONNX Runtime, model serialization, container/artifact slimming, or cold-start mitigation on Lambda/Fargate.
• Demonstrated experience deploying and monitoring ML models in production, with an understanding of model degradation, drift signals, retraining triggers, and artifact management.
• Working knowledge of deploying Python-based ML services on AWS including Lambda, ECS/Fargate, S3, IAM, and CloudWatch. Deep infrastructure expertise is not required; comfort with the AWS ML deployment stack is.
• Experience building or operating CI/CD pipelines for ML services.
• A demonstrated track record of improving production reliability through observability and disciplined deployment and maintaining healthy systems over time.
• Ability to work closely with data scientists, understanding their outputs, inheriting their work, and communicating operational decisions back to them clearly.
• Demonstrated ability to work cross-functionally with software or product engineering teams, translating ML service capabilities into integration-ready contracts and supporting those teams through adoption.
• Comfort owning work independently.
• Strong written and verbal communication.
• Willingness to perform other duties and special projects as needed/requested.
• Sensitivity and respect for racial, gender, sexual orientation, and cultural differences.
• Champions and represents Candid’s core values: We’re driven, direct, accessible, curious, and inclusive.

Spot any inaccurate information? Have a job to share? Let us know.