Senior Forward Deployed Engineer- Agentic AI

At Deloitte, Forward Deployed Engineers ( FDE ) don't just build AI solutions, they help clients turn AI ambition into enterprise-scale impact, pairing leading class engineering with pod-based delivery and vertical expertise. If you thrive at the intersection of product, engineering, problem-solving, and client impact, this role puts you at the forefront of AI transformations. As an Agentic AI Senior FDE , you will design, build, and operationalize LLM -powered systems that can reason, plan, retrieve information, use tools, and execute multi-step workflows reliably. You will work on the "thinking layer" of AI systems: agent architecture, tool orchestration, memory and context management, retrieval pipelines, evaluation, and observability. You will help shape how complex domain knowledge is transformed into usable AI behavior, with a high bar for precision, traceability, and maintainability.

Additional responsibilities include:

+ Embed with clients to identify business needs and translate high-value GenAI use cases into solutions. + Partner with leaders, product owners, architects, and engineers to align priorities and delivery. + Lead working sessions to shape solutions and drive client outcomes. + Prototype and deliver working AI solutions using industry expertise and emerging capabilities. + Contribute independently within an FDE pod while mentoring newer team members. + Build AI-enabled solutions, agentic platforms, and workflows across enterprise AI platforms. + Develop scalable AI engineering patterns, tool-use approaches, and human-in-the-loop controls. + Apply architecture decisions that balance quality, safety, latency, cost, and model risk. + Deliver production-quality code using strong practices in testing, CI/CD, logging, versioning, and documentation. + Design extensible functionality, support sprint sizing, and align solutions with senior team members. + Contribute reusable assets including code, prompt libraries, runbooks, and reference implementations. + Strong understanding of memory and context management, including context windows, retrieval-driven context assembly, persistent memory, and high-signal token selection. 4 12 + Deep understanding of how LLMs behave in practice, including strengths, failure modes, hallucination risks, reasoning limitations, latency/cost trade-offs, and evaluation methods To view full details and how to apply, please login or create a Job Seeker account