Job Description

About the Role

We are looking for an AI Engineer to help build Blue Yonder's next generation of AI-powered products.

As part of Blue Yonder's Autonomy Labs, you will work with experienced software engineers, model-training teams, data engineers, researchers, and product owners to develop agents that operate within real supply chain and retail workflows. You will contribute to the services, APIs, evaluation systems, data flows, and development environments that allow us to train, test, deploy, and continuously improve these systems.

This is a hands-on individual contributor role for a software engineer who wants to work close to the model-development lifecycle. You do not need to be a model-training researcher, but you should be curious about how model behavior, evaluation results, traces, and data can be used to build better models and more reliable products.

Your Mission

Help turn capable AI models into dependable software. You will implement and improve well-defined parts of our agent systems, contribute to technical design, and work with teammates to understand how agents behave in realistic workflows.

The challenge is not simply to make a model produce a convincing response. It is to build agents that can use tools, interact with APIs, follow business constraints, manage failures, and complete useful work reliably.

What You'll Do

• Build and maintain backend services, APIs, tools, and integrations used by AI agents and model-development workflows.
• Contribute to training and evaluation environments that represent realistic business workflows through APIs, simulations, test scenarios, and reproducible state.
• Implement pipelines and application components that collect and process agent traces, tool calls, workflow outcomes, evaluation results, and user feedback.
• Develop automated evaluations and verification checks for model and agent behavior, including task completion, tool-use correctness, and regression testing.
• Review agent traces and test results to identify failure modes, reproduce issues, and help determine whether improvements are needed in software, tools, data, prompts, or models.
• Work with model-training teams to integrate and assess new models and turn observed behavior into useful engineering feedback.
• Write clean, tested, maintainable Python code and contribute to code reviews, technical documentation, and team engineering practices.
• Improve the reliability and operability of services through logging, monitoring, automated testing, error handling, and support for safe deployments.
• Collaborate with senior engineers on system design and take ownership of scoped components from implementation through testing and operation.
• Learn quickly, share findings, and contribute practical ideas as the team develops new approaches to agent training, evaluation, and production use.

What We're Looking For

• Around three or more years of software engineering experience, or equivalent practical depth, building and maintaining production software.
• Strong Python skills and experience with backend development, including APIs, data processing, automated testing, and service integration.
• Hands-on experience developing with LLMs, AI agents, machine learning systems, or other data-intensive software.
• Familiarity with tool calling, prompt and context management, retrieval, structured outputs, or agent orchestration.
• A practical understanding of software engineering fundamentals, including maintainability, testing, version control, observability, and secure development.
• Experience working with cloud platforms such as Azure, AWS, or GCP and familiarity with containers and CI/CD workflows.
• The ability to investigate unfamiliar systems, break down problems, incorporate feedback, and deliver reliable work within a larger technical design.
• Clear communication skills and a collaborative approach to working with engineers, researchers, data specialists, and product teams.

Preferred Qualifications

• Experience creating evaluation cases, reviewing model traces, diagnosing agent failures, or testing nondeterministic systems.
• Familiarity with model-training or post-training concepts such as supervised fine-tuning, preference data, reinforcement learning, reward or verifier design, or checkpoint evaluation.
• Experience with data pipelines, dataset preparation, synthetic data, experiment tracking, or model deployment workflows.
• Experience building simulations, test harnesses, developer tools, or integrations with complex enterprise systems.
• Experience with supply chain, retail, or other environments involving multi-step workflows, permissions, business constraints, and human approvals.

What Makes This Role Different

This role offers the opportunity to develop as a software engineer at the intersection of model development and production AI. You will contribute to real systems that make agent behavior measurable and improvable, while working closely with engineers and model-training specialists on problems that span APIs, data, evaluation, and model behavior.

Our Values

If you want to know the heart of a company, take a look at their values. Ours unite us. They are what drive our success – and the success of our customers. Does your heart beat like ours? Find out here: Core Values

All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability or protected veteran status.