Senior Software Engineer, AI-First Development

PhoenixTeam is hiring three Senior Software Engineers to build and support production software for the Department of Veterans Affairs (VA). The work includes services, tools, and integrations, primarily in Python, with AI coding agents such as Claude Code serving as the primary implementation method.

This role requires an established software engineering background and the ability to work independently of AI tools when necessary. Engineers must be able to understand unfamiliar codebases, make architectural decisions, troubleshoot failures, and reason about performance, concurrency, and system behavior. Coding agents accelerate implementation, but the engineer remains responsible for the quality and reliability of the software delivered.

The Senior Software Engineer directs and verifies agent-driven development, corrects errors in generated code, and improves the workflows used to complete recurring engineering tasks. The position works closely with the Chief AI Engineer, QA & Evals, DevSecOps, Platform, and client delivery teams to move software from development into production.

Key Functions

• Design, build, and deliver production services, tools, and integrations, primarily in Python, using AI coding agents as the primary implementation method.
• Translate ambiguous requirements into defined engineering tasks, acceptance criteria, and verification steps that agents can execute.
• Own the architecture, testing, security, performance, and operational readiness of assigned software components.
• Independently review, test, and correct agent-generated code, including edge cases, failure handling, and potential regressions.
• Develop automated tests, evaluations, and acceptance checks to verify correctness and maintain software quality.
• Investigate and resolve technical issues using debuggers, profilers, logs, and underlying library source code.
• Create reusable agent skills, workflows, instructions, and runbooks for recurring engineering tasks.
• Improve agent context, prompts, and tool access while measuring development cycle time, defect rates, and review rework.
• Collaborate with QA & Evals, DevSecOps, Platform, and client delivery teams on production readiness, quality, and security requirements.
• Document technical decisions, implementation details, verification results, and handoffs for engineering and client stakeholders.
• Make implementation and code-acceptance decisions within assigned components while escalating cross-system changes and unresolved security, data, or quality concerns.

What Success Looks Like

• First 30 days: Deliver an initial production change through the standard pipeline, subject to required system access, and demonstrate effective use of the team's coding agents.
• First 60 days: Own an assigned component and establish at least one reusable engineering workflow adopted by other team members.
• First 90 days: Deliver a feature or client deliverable primarily through coding agents, supported by independently verified tests and evaluations, with review rework trending down.
• First year: Establish ownership of a defined technical area, with reusable workflows adopted as standard team practices and readiness for assessment at the next level.

Minimum Qualifications

• 5+ years of professional software engineering experience in Python, C++, Rust, Go, Java, or C#, with most of that experience gained before AI coding assistants became common (approximately 2022).
• Demonstrated experience designing, deploying, and supporting production software, including on-call support or incident response.
• Strong software engineering fundamentals in data structures, algorithms, memory management, performance, concurrency, networking, and relational databases.
• Ability to independently read, understand, modify, and troubleshoot large or unfamiliar codebases.
• Proficiency with Git, code review, automated testing, and continuous integration.
• Hands-on professional experience using AI coding agents or assistants, such as Claude Code, Cursor, GitHub Copilot, or equivalent tools, including identifying and correcting errors in generated code.
• Ability to direct coding-agent work and independently evaluate results for correctness, security, performance, and maintainability.
• Strong written and verbal communication skills, including the ability to explain technical decisions and verification results to technical and nontechnical stakeholders.

Preferred Qualifications or Skills

• Systems-level development experience in C++ or Rust involving performance-critical or safety-critical software, embedded systems, compilers, databases, trading platforms, or infrastructure.
• Advanced Python experience, including packaging, type checking, asynchronous programming, profiling, Django, or FastAPI.
• Experience with LLM APIs and managed model services, such as the Anthropic API or Amazon Bedrock, including tool use, retrieval, evaluations, and cost or latency optimization.
• Experience developing reusable agent skills, workflows, Model Context Protocol (MCP) servers, or custom tools for other engineers.
• AWS experience with ECS/Fargate, Lambda, RDS, IAM, or comparable cloud services.
• Experience delivering software in federal government, mortgage, financial services, or other regulated environments.
• Client-facing software delivery or technology consulting experience.
• Experience mentoring junior software engineers.

Application Requirements

Applicants must provide evidence of previous software engineering work through one of the following:

• Code samples: Links to personally authored code in GitHub, GitLab, or comparable repositories, including contributions predating 2022 in one or more of the required programming languages.
• Project descriptions: For work that cannot be shared publicly, descriptions of two or three completed projects covering approximate dates, the problem addressed, personal responsibilities, programming languages, architecture, scale, and outcomes.

Applicants must also provide a brief example of professional AI-assisted software development, identifying the tools used, the work completed, what they personally reviewed or corrected, and the result.

Do not submit proprietary, confidential, classified, or otherwise restricted code or information. Project descriptions should include only information authorized for disclosure.

Submissions will be evaluated for demonstrated engineering ownership, technical design, testing practices, and the ability to independently verify AI-generated work.

Position Details

• Employment type: Full-Time, W2 employee.
• Work arrangement: Remote-first within the United States, with some travel to client sites.
• Client: Department of Veterans Affairs (VA).
• Openings: 3.
• Eligibility: Must be able to obtain and maintain a favorable federal Public Trust determination.

About PhoenixTeam

PhoenixTeam partners with organizations to deliver real results across GenAI, data, product, and technology. We help clients modernize legacy systems, redesign complex workflows, and move practical AI and digital solutions into production in regulated environments.

PhoenixTeam is an Equal Opportunity Employer. We consider qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, genetic information, protected veteran status, or any other status protected by applicable law.