Context Engineer

USAA•San Antonio, TX

4d•$77,120 - $147,390

About The Position

USAA is seeking a Context Engineer to work at the intersection of business strategy, data science, and engineering. The role shapes how business context, data, and logic are structured to support generative AI, large language models, and autonomous agents. The engineer architects prompts, workflows, and agent behaviors, iterates on AI solutions, optimizes model outputs, and supports safe, compliant agent control design. The position works with cross-functional business, data science, and engineering teams to translate business objectives into AI workflows; design and evaluate prompt architectures and retrieval-augmented generation pipelines; and engineer and monitor AI agents. It also involves assessing model behavior and output quality, managing AI-related risks, documenting systems for review and audits, communicating findings to stakeholders, and adopting emerging AI technologies and tools. USAA's mission is to empower members to achieve financial security through competitive products, exceptional service, and trusted advice, with a focus on serving the military community and their families. The salary range is $77,120–$147,390, with actual pay varying by location, experience, and market data. Employees may be eligible for performance-based incentives. USAA does not provide visa sponsorship for this role. Applications are accepted on an ongoing basis until the position is filled.

Requirements

Bachelor's degree in Computer Science, Artificial Intelligence, Data Science, Software Engineering, Information Systems, or a related technical or quantitative field, or four years of relevant education and/or technical experience.
At least 2 years of hands-on experience with modern AI/ML technologies, data analytics, or software development involving LLM applications, prompt engineering, or workflow automation; alternatively, an advanced degree in Computer Science, AI, Data Science, or a related quantitative or technical discipline.
Demonstrated experience with LLMs, prompt engineering patterns such as Chain-of-Thought and few-shot prompting, and major foundation model APIs or open-weights models.
Practical familiarity with AI orchestration frameworks and toolsets such as LangChain, LlamaIndex, Semantic Kernel, or custom agent frameworks, and with vector databases.
Proficiency in Python and modern data manipulation tools, with the ability to query structured and unstructured data sources using SQL and APIs to supply context to AI pipelines.
Solid understanding of AI risk management, guardrails, security best practices such as mitigating prompt injection and protecting data privacy, and agent control design.
Must not require immigration support now or in the future; USAA does not provide visa sponsorship for this role.

Responsibilities

Collaborate with business, data science, and engineering teams to translate business objectives into AI and agentic workflows.
Design, test, and optimize prompt architectures, context windows, and retrieval-augmented generation (RAG) pipelines to improve LLM performance and accuracy.
Engineer and monitor autonomous or semi-autonomous AI agents, participating in iterative design and continuous refinement.
Use qualitative and quantitative evaluation frameworks to assess model behavior, hallucinations, latency, and output quality against business use cases.
Follow risk, safety, and agent control design standards to support ethical, aligned, and compliant AI deployment.
Synthesize AI outputs and multimodal data insights into actionable recommendations and communication materials for peer review and stakeholder alignment.
Document system instructions, prompt templates, agent logic, and testing frameworks for peer reviews and compliance audits.
Explore and adopt emerging generative AI technologies, frameworks, and tools to enhance solution strategies.
Present findings and demonstrate agent capabilities to influence mid-level management, gain stakeholder buy-in, and accelerate AI adoption.
Identify, measure, monitor, and control risks related to AI operations, data leakage, and automated business activities.