Generative AI Applications Engineer (Agents & RAG)

Accenture Federal Services•Washington, DC

5h•$103,200 - $203,400

About The Position

Accenture Federal Services is hiring a Generative AI Applications Engineer to build secure, reliable, and scalable generative AI applications for confidential U.S. federal programs. This hands-on role spans agentic workflows, retrieval-augmented generation (RAG), prompt and policy design, LLM evaluation, and platform integration; model training is not required. The engineer will own the path from evaluating use cases through production deployment and ongoing operational excellence, partnering with product, security, data, and SRE teams to deliver features safely and at scale. The role emphasizes production GenAI systems designed for mission data and constrained environments, with attention to hallucination rates, latency, reliability, safety, and cost. Work includes implementing managed cloud AI services and retrieval pipelines, evaluating and integrating models, adding observability and guardrails, and delivering reusable platform components. Solutions may need to operate in hybrid, restricted, or air-gapped environments using Zero Trust principles and audit-ready controls. Accenture Federal Services describes itself as a technology company within global Accenture, dedicated to serving the U.S. federal government. It offers a collaborative community and opportunities for hands-on experience, certifications, and industry training. The stated base pay range for specified locations is $103,200–$203,400 USD. Applications are accepted on an ongoing basis with no fixed deadline.

Requirements

Active U.S. citizenship.
Experience building and deploying a production generative AI application, such as a chatbot, copilot, assistant, or enterprise AI application.
Hands-on experience with LLMs such as GPT, Claude, Llama, Gemini, or Mistral through APIs or self-hosted deployments.
Experience designing and implementing RAG solutions using embeddings, vector databases, and semantic search for enterprise or mission data.
Experience shipping AI applications with frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, DSPy, or similar.
Strong Python development skills for building and integrating AI/ML applications.
Experience owning AI solutions through the full production lifecycle, including deployment and operational support.

Nice To Haves

Experience integrating leading cloud AI services or on-premises inference stacks.
Background in LLM evaluation, prompt authoring and testing, A/B experimentation, and LLM operations.
Responsible AI expertise, including privacy, security, bias, transparency, human-in-the-loop practices, and data governance.
Experience implementing tool-using agents for API integration and external data access.
Experience with containerization and orchestration tools such as Docker, Kubernetes, or VMware, and scripting or automation using Linux Bash or PowerShell.
Prior work in regulated or secure environments, including ATO, STIGs, or Zero Trust, while delivering quickly.
Familiarity with NVIDIA AI Foundations, OpenAI ChatGPT, and AI-assisted development tools such as Cursor, Windsurf, or Claude.
Contributions to internal frameworks or open source, and experience mentoring engineers.
Clear communication with engineers, product managers, and security or compliance stakeholders.

Responsibilities

Build and deploy mission-grade generative AI applications, including agentic workflows and RAG systems tailored to mission data and environments.
Design agent orchestration, task decomposition, tool use, guardrails, and recovery or fallback strategies using frameworks such as LangChain, LlamaIndex, and Semantic Kernel.
Integrate managed AI platforms and services, including AWS Bedrock, Azure OpenAI, Google Vertex AI, Amazon Kendra, and services such as Document AI, Gemini, and Gemma; model training is not required.
Compare LLMs for quality, safety, latency, and cost; author and test prompts and policies; and deploy with observability and safe rollback or fallback.
Build retrieval pipelines and vector search, including data preparation, chunking, metadata handling, and information-retrieval evaluation.
Instrument metrics, logs, and traces; run A/B experiments; maintain incident playbooks; and implement safety and compliance guardrails.
Define AI service-level indicators and objectives for quality, latency, safety, and cost; participate in on-call operations and postmortems; reduce MTTR; and optimize token usage and spending.
Develop reusable SDKs, CI/CD templates, Terraform or other infrastructure-as-code modules, and evaluation harnesses for multiple mission teams.
Deliver solutions in hybrid, restricted, or air-gapped environments using Zero Trust principles and audit-ready controls.
Own AI solutions across the production lifecycle, including deployment and operational support.