AI / ML Engineer

nelnet; United States; On-site; USD 120,000 - 180,000; Full time

Job summary

Nelnet is seeking an AI / ML Engineer to advance agentic systems, maintain deployed agents, and design new capabilities across the business. You will partner with Data Scientists, Product, and Software Engineers to move solutions from prototype to production.

The role covers building LLM-powered agents, improving retrieval, ensuring safety and governance, and scaling AI pipelines on AWS with Terraform and CI/CD.

Qualifications

• Strong programming skills in Python.
• Working knowledge of AWS services for AI workloads (Bedrock, Lambda, ECS/Fargate, S3, OpenSearch, Step Functions).
• Familiarity with agent frameworks and orchestration tooling.
• Practical prompt and context engineering with an evaluation-driven approach.

Responsibilities

• Agent Development: Design, build, and deploy LLM-powered agents to solve concrete business problems.
• Operating the Existing Fleet: Maintain health of agents in production; monitor, diagnose, and upgrade.

Skills

Python

AWS

Agent frameworks

Prompt engineering

Context engineering

Evaluation

Tools

Terraform

CI/CD

Bedrock

Lambda

ECS/Fargate

S3

OpenSearch

Step Functions

Job description

Nelnet is a diversified and innovative company committed to enriching lives through the power of service as a student loan servicer, professional services company, consumer loan originator and servicer, payments processor, renewable energy solutions, and K-12 and higher education expert. For over 40 years, Nelnet has been serving its customers, associates, and communities.

The perks of working at Nelnet go beyond our benefits package. When you join the Nelnet team, you're part of a community invested in the success of each individual. That support comes through in our work, as we are united by our mission of creating opportunities for people where they live, learn, and work.

At Nelnet, we believe in the transformative power of data and technology to drive business innovation. Our AI / ML Engineer role sits at the center of that work: building, operating, and scaling the AI agents that automate processes, surface insights, and help our teams make better decisions faster. If you're passionate about taking agentic systems from prototype to dependable production capability, we invite you to join our team and help shape how AI gets used across our business.

This posting covers multiple levels within the AI / ML Engineer role. We are currently seeking candidates at the equivalent of a Level I or II AI / ML Engineer. The responsibilities and qualifications below describe the full scope of the role; expectations for depth, autonomy, and scope of ownership will vary by level. Level and corresponding compensation are determined during the interview process based on demonstrated experience. We encourage you to apply if you meet the core qualifications, even if you don't match every item listed.

Role Overview

As an AI / ML Engineer at Nelnet, you'll be at the intersection of applied AI and software engineering. Your primary focus will be twofold: keeping our existing fleet of deployed agents healthy, evaluated, and improving over time, and designing new agentic capabilities that extend what those systems can do. Beyond building, you'll help expand the agentic approach across the organization, partnering with business stakeholders to identify high-value use cases, establishing reusable patterns and guardrails, and raising the bar for how teams evaluate and ship AI systems. You'll collaborate closely with Data Scientists, Product, Software Engineers, and business partners to move solutions from concept to production.

Job Responsibilities

1. Agent Development : Design, build, and deploy LLM-powered agents to solve concrete business problems. This includes tool use, multi-step reasoning, orchestration, and human-in-the-loop patterns.
2. Operating the Existing Fleet : Own the day-to-day health of agents already in production: monitor behavior, diagnose failures, tune prompts and tooling, and manage model and dependency upgrades without regressing quality.
3. Evaluation : Build and maintain evaluation suites for agent systems, including offline test sets, LLM-as-judge scoring, regression testing, and online metrics.
4. Observability and Monitoring : Instrument agents end to end, including traces, tool calls, token usage, latency, cost, and outcome quality and act on what the data shows.
5. Context and Retrieval Engineering : Design retrieval and context strategies (RAG, structured data access, caching, chunking, ranking) that give agents the right information at the right time.
6. Tooling and Integrations : Build and maintain the tools, APIs, and connectors agents rely on, ensuring safe and reliable interaction with internal systems and data.
7. Scalability and Infrastructure : Design and implement scalable AI pipelines and services on AWS, using infrastructure as code (Terraform) and CI/CD to automate deployment and maintenance.
8. Guardrails and Responsible AI : Implement safety controls, input/output validation, access boundaries, and audit trails appropriate to a regulated environment.
9. Expanding Agentic Adoption : Partner with teams across the organization to identify where agents add real value, prototype quickly, and turn one-off wins into reusable frameworks and standards.
10. Documentation : Maintain clear documentation of agent architectures, prompts, tool contracts, data flows, evaluation results, and known limitations.
11. Innovation : Track the fast-moving foundation model and agent tooling landscape, and bring what's genuinely useful into our stack.
12. Mentorship : Provide guidance to team members on agent design, evaluation practices, and applied AI best practices.

Key Competencies

1. Strong programming skills in Python.
2. Working knowledge of AWS services for AI workloads (e.g., Bedrock, Lambda, ECS/Fargate, S3, OpenSearch, Step Functions).
3. Familiarity with agent frameworks and orchestration tooling, and sound judgment about when a framework helps versus when to build directly.
4. Practical prompt and context engineering skill, with an evaluation-driven approach to improving them.
5. Experience wi