AECOM is seeking an Agentic AI Architect to architect, design, build, and optimize enterprise-grade AI platforms, focusing on modern AI frameworks and AI-driven platform development to ensure scalability, security, and performance in production systems.
Requirements
- Hands-on architect abilities to collaborate with current Agile SCRUM team.
- Ability to work with little guidance on POCs and forward-looking research on upcoming AI technologies.
- Strong front-end experience with Angular (latest versions preferred).
- Expertise in SQL Server including complex queries, performance tuning, and data modeling.
- Hands-on experience with Azure Cloud services including monitoring and troubleshooting Cloud issues.
- Practical experience with Agentic Workflows, AI Agents, MCP, API Management (APIM), Semantic Kernel, Kernel Memory, AI Foundry, or similar AI frameworks.
- Familiarity with LLM integration and Retrieval Augmented Generation (RAG) architectures.
Responsibilities
- Define and evolve the enterprise AI platform architecture, breaking down business and functional requirements into scalable, secure, and maintainable technical designs.
- Establish the long-term architecture runway for AI products, owning POCs, design spikes, and forward-looking evaluations of emerging technologies—including agentic architectures, orchestration frameworks, and tool-use patterns.
- Architect and implement enterprise-grade RAG (Retrieval Augmented Generation) pipelines, including vector search, embeddings, knowledge stores, and LLM orchestration.
- Design and operationalize agentic workflows, including multi-agent collaboration, tool-calling agents, planner/executor patterns, and automated reasoning loops.
- Build and integrate production-grade AI agents capable of interacting with external tools, APIs, enterprise systems, and knowledge sources.
- Utilize the Model Context Protocol (MCP) to expose internal tools, datasets, and enterprise APIs to LLMs in a secure, governed manner.
- Integrate and operationalize multiple LLM providers (Azure OpenAI, OpenAI, Anthropic, open-source models) into production systems.
Other
- This position offers flexibility for hybrid work schedules to include both in-office presence and telecommute/virtual work, to be based in Houston, TX or Dallas, TX.
- Partners with enterprise architects, product managers, developers, and data engineers in an Agile/Scrum environment to ensure architecture aligns with product strategy.
- Provide hands-on mentorship on AI development patterns—including agent-building, RAG troubleshooting, prompt design, vector search patterns, and orchestration frameworks.
- Sponsorship for US work authorization is not available for this position, now or in the future
- Relocation assistance is not available for this position
