Agentic AI and LLM Applications Software Development Engineer, Senior
Booz Allen Hamilton7 days ago
Base Salary
$87k - $198k/yr
Responsibilities
- Design and build multi-agent GRACE workflows for reasoning, planning, memory, tool use, collaboration, and task handoffs.
- Build and maintain the tool-calling layer, including schemas, error handling, retries, result formatting, and MCP client-side tool discovery and invocation.
- Own application-layer LLM orchestration, including prompts, context management, model selection, response parsing, token efficiency, and integrations across GPT, Claude, and Gemini.
- Build RAG capabilities, LLM quality-evaluation pipelines, regression tests, safety checks, and prompt/model experiments.
- Translate ambiguous product requirements into technical designs and ship backend application logic and frontend-facing API contracts end-to-end.
- Prototype and iterate on agentic features using real user behavior and collaborate with product, UX, applied science, and operations.
- Ensure production quality through testing, graceful degradation, tracing, logging, metrics, fallback logic, guardrails, privacy, security, and compliance.
- Define and monitor application-level SLOs for tool-call success, response quality, and latency.
- Mentor and unblock engineers, participate in design reviews, and communicate technical decisions clearly.
Requirements
- 7+ years of software engineering experience, including building and operating production systems.
- Experience owning and shipping complex products end-to-end in high-velocity environments.
- Experience with Python and at least one other backend language.
- Experience building and operating systems on AWS, GCP, or Azure.
- Experience with containerization and CI/CD pipelines.
- Knowledge of modern backend frameworks, asynchronous patterns, distributed systems, APIs, data pipelines, and software design patterns.
- Bachelor’s degree in Computer Science or Software Engineering.
- Strong communication, empathy, ownership, self-direction, urgency, and feedback skills.
- Completion of Booz Allen’s required skills assessment and achievement of the minimum proficiency threshold.
- Preferred: production LLM systems involving tool calling, RAG, multi-step reasoning, context management, multi-agent A2A architectures, orchestration frameworks, and LLM evaluation pipelines.
- Preferred: startup or early-stage product development, customer-facing AI platforms or developer tools at scale, security-conscious engineering, regulated-domain experience, MCP client-layer knowledge, token economics, and prompt engineering across model families.
Benefits
- Health, life, disability, financial, and retirement benefits are offered through Booz Allen benefit programs.
- Paid leave, professional development, tuition assistance, work-life programs, dependent care, and recognition awards are available.
- Regular full-time and part-time employees working at least 20 hours per week are generally eligible for benefit programs, with select offerings for others.
- The role may be remote, hybrid, or onsite according to the posting and may require work at Booz Allen or customer facilities; virtual employees are generally expected to use cameras during meetings.
- Candidates must complete identity verification involving biometrics and remain on camera during interviews and assessments.
Tech Stack
Categories
About Booz Allen Hamilton
Whether at the border, up in space, or on the battlefield, we build the advanced technology that makes America stronger, faster, and safer. It’s who we are and what we do. It’s in our code.