Senior AI Platform Engineer (Agentic LLM Orchestration)

Advanced Resource Technologies, Inc.

Washington (District of Columbia)

On-site

USD 140,000 - 200,000

Full time

2 days ago

Be an early applicant

Get more replies from employers

Send a job-specific resume in minutes.

Job summary

Advanced Resource Technologies, Inc. is seeking a Senior Software Development Engineer III to support ARPA-H’s GRACE development team.

This role focuses on building agentic AI systems, LLM orchestration, tool-calling, and AI-powered product development for GRACE, ARPA-H’s production assistant. You will design multi-agent workflows, manage LLM orchestration, and implement tool-calling infrastructure while collaborating with product, UX, and operations to deliver reliable, observable solutions

Qualifications

Bachelor’s or Master’s degree in Computer Science, Software Engineering, or related field, or equivalent practical experience.
7+ years of professional software engineering experience building and operating production systems.
Proven experience building and shipping complex products in high-velocity environments.
Strong proficiency in Python and at least one additional backend programming language.
Experience with algorithms, APIs, data structures, and software design patterns.
Experience building systems on major cloud platforms including AWS, GCP, or Azure.
Experience with containerization and CI/CD pipelines.
Hands-on experience building production systems on top of LLMs including tool-calling, RAG, multi-step reasoning, and context management preferred.
Experience with multi-agent orchestration frameworks and MCP integrations preferred.
Familiarity with prompt engineering, token economics, and LLM evaluation pipelines preferred.
Experience building customer-facing AI platforms or developer tools at scale preferred.
Background in security-conscious engineering and responsible AI guardrails preferred.
Strong communication, collaboration, and mentoring skills.
Ability to work collaboratively with cross-functional teams.
Experience in healthcare, life sciences, or regulated environments preferred.

Responsibilities

Design and build GRACE’s core agentic workflows including multi-step reasoning, planning, memory, and tool usage
Implement and evolve A2A communication patterns enabling collaboration across single and multi-agent systems
Build and maintain tool-calling infrastructure including tool definitions, schemas, retry logic, and result formatting
Develop MCP client-side integrations enabling GRACE agents to discover and invoke tools
Design multi-agent workflows that are observable, reliable, and production-ready
Own LLM orchestration including prompt construction, context management, model selection, and response parsing
Build and maintain RAG capabilities including query formulation, citation grounding, ranking, and hallucination mitigation
Implement prompt engineering patterns and system prompts across OpenAI GPT, Anthropic Claude, and Google Gemini
Manage context window budgets including truncation, summarization, and pagination logic
Build evaluation pipelines for grounding assessment, regression testing, safety checks, and A/B experimentation
Develop new AI-powered capabilities end-to-end from backend application logic to frontend API contracts
Rapidly prototype new agentic AI features and iterate based on user feedback and experimentation
Collaborate with product, UX, operations, and applied science teams to deliver user-focused solutions
Instrument workflows with tracing, logging, and metrics to ensure observability and reliability
Define and monitor application-level SLOs including tool success rates, response quality, and latency
Implement fallback and guardrail logic for AI services and model outputs
Write production-quality code that is readable, tested, documented, and maintainable
Participate in design reviews and mentor engineers across the team
Ensure privacy, security, and compliance across all application logic and data handling