Agentic AI Systems Engineer

Applied Materials•Santa Clara, CA

23h•Onsite

About The Position

The Agentic AI Systems Engineer is responsible for building the agentic infrastructure that enables next generation of GenAI applications for Applied Products. The role bridges AI, software infrastructure, and domain engineering to ensure that agentic systems are secure, deterministic, and drive velocity and productivity for our products. What You Will Build Skill Pack Framework Architect and ship the end-to-end skill framework, including authoring, versioning, distribution, and runtime loading across Java, Python, C++, and web-based codebases. Skills are the core differentiator—encapsulated expertise that travels with the agent. Multi-Agent Runtime Design and implement a multi-agent execution environment with persistent background agents, swarm coordination, and cross-agent context sharing. Enable agents to operate continuously and in parallel, not as one-shot invocations. Agent-Aware Code Intelligence Develop deep code understanding systems such as semantic code search, dependency-aware context modeling, and structural codebase analysis. Move beyond text-based retrieval toward true program-level reasoning. Key Responsibilities : Architect, design and implement the agent runtime and orchestration platform that enables AI assisted development workflows across Applied products. Act as a bridge between AI research capabilities and real production software needs, translating emerging agentic patterns into deployable infrastructure. Own the end-to-end lifecycle of agentic systems, define core abstractions for agents, tools, memory, planning, validation, and execution, ensuring reuse and consistency across teams. Partner with teams and develop the MCP platform, unifying code, documentation, and developer lifecycle MCPs into a single, dependable context infrastructure. Design mechanisms for secure tool access, API allow listing, and controlled execution against internal systems. Establish evaluation frameworks to track correctness, latency, cost, safety events, and human override frequency. Build instrumentation for logging, metrics, and tracing of agent behavior across sessions and workflows. Provide technical leadership and mentorship to engineering teams adopting agent based workflows.

Requirements

Strong system level thinking with the ability to design end-to-end architectures spanning agents, tools, memory, governance, and observability.
Experience designing agent runtimes or orchestration layers (not just using frameworks).
Solid system design expertise for highly reliable, observable, developer-facing services and agentic systems.
Strong proficiency in at least one major programming language (e.g., Java, Python, C++) with hands-on experience integrating AI/ML models into developer workflows.
Proven experience building and operating developer tooling or platform infrastructure (SDKs, CLIs, APIs, runtimes) in large-scale engineering environments.
7+ years of industry relevant experience

Nice To Haves

Experience with advanced agentic patterns (multi‑agent workflows, planner/critic loops).
Hands on exposure to on prem LLM deployment and optimization.
Practical knowledge of RAG and vector search systems for grounding AI outputs
Background in security first or high reliability software environments

Responsibilities

Architect, design and implement the agent runtime and orchestration platform that enables AI assisted development workflows across Applied products.
Act as a bridge between AI research capabilities and real production software needs, translating emerging agentic patterns into deployable infrastructure.
Own the end-to-end lifecycle of agentic systems, define core abstractions for agents, tools, memory, planning, validation, and execution, ensuring reuse and consistency across teams.
Partner with teams and develop the MCP platform, unifying code, documentation, and developer lifecycle MCPs into a single, dependable context infrastructure.
Design mechanisms for secure tool access, API allow listing, and controlled execution against internal systems.
Establish evaluation frameworks to track correctness, latency, cost, safety events, and human override frequency.
Build instrumentation for logging, metrics, and tracing of agent behavior across sessions and workflows.
Provide technical leadership and mentorship to engineering teams adopting agent-based workflows.