Manager – Agentic AI Engineering (Forward Deployed) Job Posting

Manager – Agentic AI Engineering (Forward Deployed)

Location: Flexible / Hybrid | Reports to: Senior Manager, Application Security & AI Engineering | Experience: 7+ years hands-on engineering

About the Role

We are hiring a Manager, Agentic AI Engineering to operate as a forward deployed engineer: embedded with clients and accountable for scoping, designing, building, and deploying agentic AI solutions into production. This is a hands-on engineering role, not an advisory one. You will write and ship code, build agent harnesses, and make model-agnostic architecture decisions across frontier and open-weight models. You report to the Senior Manager, and work daily in AI-assisted IDEs, source control, and CI/CD pipelines.

Key Responsibilities

** Scope:** Work directly with client engineering and business stakeholders to identify high-value agentic use cases and define scope, success criteria, and acceptance tests.

** Design:** Produce model-agnostic agentic architectures covering agent topology, tool and data integrations, memory and context strategy, model selection, and guardrails.

** Build:** Write production code for agents and harnesses: orchestration, tool calling, Model Context Protocol (MCP) servers and clients, context management, evaluations, observability, and human-in-the-loop checkpoints.

** Deploy:** Move solutions from prototype to pilot to production through Git-based workflows and CI/CD pipelines with automated tests and evaluation gates.

** Own AI-generated code:** Use AI coding agents as primary development tools, and review, test, and take ownership of all AI-generated code before merge.

** Secure by default:** Apply sandboxing, least-privilege credentials, tool-access boundaries, and prompt-injection defenses to every agent you deploy.

** Field insights:** Capture reusable patterns, accelerators, and lessons learned, and brief the Senior Manager and pursuit teams on what works in production.

** Mentor:** Coach engineers and consultants on agentic engineering, harness design, and AI-assisted development.

Required Qualifications (Must-Have)

** Engineering foundations:** 7+ years of hands-on software engineering (backend, platform, or ML engineering) with code shipped to production in Python, TypeScript, Java, Go, or C#, including:

• ** AI-assisted development:** Daily use of VS Code or another IDE (e.g., JetBrains) with at least one of Claude Code, Cursor, GitHub Copilot (including Copilot CLI), OpenAI Codex, or Windsurf.

• ** Source control:** Proficient with Git (branching, rebasing, pull requests, code review) on GitHub, GitLab, Bitbucket, or Azure Repos.

• ** CI/CD:** Built and maintained pipelines hands-on in at least one of Concourse, Bamboo, GitHub Actions, GitLab CI/CD, Jenkins, Azure DevOps Pipelines, CircleCI, AWS CodePipeline/CodeBuild, Google Cloud Build, Harness, Tekton, or Argo CD.

** Agentic delivery:** Deployed 2 or more agentic AI solutions to production or a client pilot, with personal ownership of the design and build, not only oversight. Built with at least two model providers (e.g., OpenAI, Anthropic, Google Gemini, or open-weight models via Amazon Bedrock, Vertex AI, Azure AI Foundry) and at least one agent framework (e.g., LangGraph, OpenAI Agents SDK, Claude Agent SDK, Microsoft Agent Framework, Google ADK).

** Agentic architecture and harness design:** Built at least one agent harness from its components: orchestration loop, tool calling, MCP integration, context and memory management, evaluations, and guardrails. Able to explain and defend design trade-offs, including single vs. multi-agent, deterministic workflow vs. autonomous agent, model choice by cost, latency, and quality, and failure and escalation handling.

** MCP understanding:** Solid understanding of the Model Context Protocol (MCP), including how MCP servers and clients are designed, secured, and used to connect agents to tools and data.

** Software supply chain security:** Working knowledge of securing the software supply chain, including generating and consuming SBOMs (SPDX, CycloneDX), SCA scans, dependency and vulnerability management (CVEs, CVSS, VEX), 0-day response and rapid remediation, artifact signing and provenance (e.g., SLSA, Sigstore), and securing build pipelines and third-party dependencies, including AI/ML models and MCP servers.

** Client engagement:** Has led technical discovery and solution sessions directly with client stakeholders and can take an idea from whiteboard to working demo. Adapts quickly and has the flexibility to keep pace with changing client requests and needs.

** Program health and leadership:** Able to maintain overall program health and lead teams, building and leading high-performing teams. Proactively communicates with, and notifies, the Senior Manager and leadership of blockers or risks as they emerge.

** Thought leadership and research:** Keep up to date with research and market trends, and stays ahead of the fast-moving evolution and advancements in agentic AI and the broader field. Has the ability to create industry-leading perspectives (POVs), and to support leadership in identifying, defining, and building service offerings that meet real market needs.

Preferred Qualifications (Nice-to-Have)

** Application Security:** Threat modeling (STRIDE, MITRE ATLAS), SAST and SCA tooling (Semgrep, CodeQL, Snyk), and the OWASP Top 10 for LLM and agentic applications.

** AI Security:** Red teaming, prompt-injection testing, RAG security, agent security, and guardrails.

** Evaluation & observability:** LangSmith, Langfuse, Arize Phoenix, Promptfoo, or OpenTelemetry-based tracing.

** Agent interoperability:** Authoring MCP servers and working with Agent2Agent (A2A) protocol.

** Cloud & platform:** Azure, AWS, Google Cloud, Docker, Kubernetes, and Terraform, with policy-as-code (OPA).

** RAG & data:** Vector stores (e.g., pgvector, Pinecone, Azure AI Search), embeddings, and retrieval evaluation.

** Domain:** Consulting or forward deployed engineering experience in financial services or other regulated industries.

Certifications (At Least One Required)

Candidates must hold at least one current certification, or have completed the equivalent training, from the Microsoft, AWS, or SANS/GIAC options below. Any one is sufficient; holding more is a plus.

• ** AI:** Microsoft Azure AI Apps and Agents Developer Associate (AI-103), Machine Learning Operations Engineer Associate (AI-300), or Multi-Agent AI Solutions Expert (AI-500); AWS Certified Generative AI Developer – Professional, Machine Learning Engineer – Associate, or AI Practitioner; or SANS/GIAC SEC545: GenAI and LLM Application Security (GAIPS), SEC595: Applied Data Science and AI/Machine Learning for Cybersecurity Professionals, SEC535: Offensive AI – Attack Tools and Techniques, or SEC598: AI and Security Automation for Red, Blue, and Purple Teams.

• ** Cloud:** Microsoft Azure AI Cloud Developer Associate (AI-200) or Azure Administrator Associate (AZ-104); AWS Certified Developer – Associate or DevOps Engineer – Professional; or SANS/GIAC SEC488: Cloud Security Essentials or SEC510: Cloud Security Engineering and Controls.

• ** Architecture:** Microsoft Azure Solutions Architect Expert (AZ-305) or Agentic AI Business Solutions Architect Expert (AB-100); AWS Certified Solutions Architect – Associate or Professional; or SANS/GIAC SEC549: Cloud Security Architecture.

What You’ll Bring

A builder’s track record: you have shipped agents that clients use, and you can show the code, the harness, and the results. You will help clients move agentic AI from pilot to production, securely and independent of any single model vendor.

**Differentiator: **Two or more production agentic solutions you built hands-on using Claude Code, Cursor, or GitHub Copilot, plus working knowledge of AppSec or Secure AI SDLC.

• Travel Requirements:

• Location – Philadelphia, New York, Charlotte, Dallas

• Travel – Remote (10-15% client travel for key meetings)

$95-$100/hr

Manager – Agentic AI Engineering (Forward Deployed) | Job Posting