Forward Deployed AI Engineer

The Forward Deployed AI Engineer is a hands-on builder responsible for designing, building, and deploying agentic AI capabilities and workflows across AGS's business operations and software development processes. This role works at the frontier of what's possible with large language models and multi-agent systems — building the AGS One / RevMax sales copilot, regulatory monitoring agent, AI-assisted QA testing agent, and SDLC automation tools that make up the core of AGS's AI transformation program. Agents are standardized on Azure AI Foundry as the primary runtime so they are reusable, governed, and discoverable across teams rather than one-off builds. Responsibilities Build and deploy agentic AI capabilities and workflows — multi-step agents that reason, use tools, and take autonomous actions, built with LangGraph and hosted on Azure AI Foundry Agent Service (AGS's standard runtime); use CrewAI, AutoGen, or equivalent frameworks only where a workflow genuinely outgrows Foundry Own AI tooling deployment — Cursor, Factory.AI, GitHub Copilot, Claude Code — integrating AI coding tools into AGS's development workflows and measuring productivity impact Build and deploy the AGS One / RevMax sales copilot — AI agents that query Salesforce, pull game performance data, generate proposals, and support account managers across the revenue lifecycle Build and deploy the regulatory monitoring agent — an agent that monitors regulatory feeds, classifies relevant changes, assesses business impact, and routes alerts to the right teams Build AI-assisted QA testing workflows — automated test execution, pre-certification compliance checking, and test report generation for game builds, alongside OmniBet-based porting tools Own AI pipeline orchestration — design and maintain the workflows connecting LLMs, data sources, vector databases, and downstream applications Implement RAG systems — retrieval-augmented generation using Azure AI Search (or equivalent vector store) for agents that need to reason over AGS's proprietary data (game portfolio, performance history, regulatory database) as it is surfaced through Fabric/OneLake Design agentic deployment infrastructure — containerization, API endpoints, versioning, and monitoring for all production AI agents, registered in the Azure AI Foundry / Entra agent registry Build and consume MCP server integrations — connect agents to internal APIs, data sources, and tools via Model Context Protocol servers so capabilities are reusable across agents rather than rebuilt each time Implement evaluation frameworks — measure agent accuracy, reliability, latency, and cost; build dashboards that surface these metrics to the Head of AI Stay current on agentic AI developments — evaluate new LLM capabilities, orchestration frameworks, and deployment patterns; bring relevant advances into AGS's platform Skills/Requirements 4–8 years of software engineering experience , with the last 1–3 years specifically focused on LLM application development and agentic AI systems Production agentic AI experience — has built and shipped AI agents to production, not just prototype demos; ideally using LangGraph (AGS's primary framework), or LlamaIndex, CrewAI, or AutoGen Strong Python development skills — production-quality Python for agent development, API integration, data processing, and deployment LLM API experience — hands-on experience with Anthropic Claude, OpenAI, or equivalent LLM APIs (via Azure AI Foundry or direct) including function calling, tool use, and structured output RAG system experience — has built retrieval-augmented systems with vector databases (e.g., Azure AI Search, Pinecone, pgvector); understands embedding models, chunking strategies, and retrieval optimization Prompt engineering skills — systematic approach to prompt design, versioning, and evaluation; understands chain-of-thought, few-shot, and structured output prompting API and systems integration — experience integrating with Salesforce, databases, REST APIs, and enterprise systems from within agent workflows Monitoring and observability — experience instrumenting AI applications for production monitoring including latency, cost, and quality metrics Bachelor's degree in Computer Science, Engineering, or equivalent practical experience Preferred Experience with SDLC tooling deployment and CI/CD integration Familiarity with gaming domain — game mechanics, certification processes Experience with responsible AI practices, guardrails, content-safety baselines, and audit logging Familiarity with Microsoft Copilot Studio, Azure AI Foundry agent registry, or similar agent governance/deployment patterns Experience building or consuming Model Context Protocol (MCP) servers Note: All offers are contingent upon successful completion of a background check #J-18808-Ljbffr