Principal AI Systems Engineer — C++ / Applied AI

Adobe Adobe · Enterprise · San Jose, CA +4 · Remote

Principal AI Systems Engineer with deep C++ expertise to build next-generation AI-enabled product and platform capabilities. This role focuses on designing and building native infrastructure, service integration layers, evaluation systems, and reliability mechanisms for AI-powered features within complex software products. It requires a strong C++ background with practical AI fluency, focusing on production integration, reliability, observability, and security of AI systems.

What you'd actually do

  1. Design and build native C++ infrastructure that connects complex product codebases to AI-powered services, agents, and model-backed workflows.
  2. Build high-quality C++ components for performance-sensitive, cross-platform environments.
  3. Design systems that make AI features measurable, debuggable, and production-ready.
  4. Act as a technical lead across teams building AI-powered product infrastructure.
  5. Set engineering direction in ambiguous and fast-moving technical areas.

Skills

Required

  • 10+ years of professional software engineering experience
  • significant depth in C++
  • modern C++ design
  • memory management
  • concurrency
  • API design
  • debugging
  • systems-level performance
  • cross-platform software development
  • integrating AI, LLMs, agents, or model-backed systems into production
  • AI system failure modes
  • designing reliable interfaces between AI systems and deterministic software systems
  • systems architecture
  • tradeoff analysis
  • technical strategy
  • architecture documentation
  • testing at scale
  • unit tests
  • integration tests
  • CI validation
  • regression testing
  • quality gates
  • technical leadership
  • influence across teams
  • written and verbal communication

Nice to have

  • frontier model APIs
  • tool/function-calling interfaces
  • agentic workflows
  • AI orchestration systems
  • AI evaluation frameworks
  • automated scoring systems
  • human-in-the-loop quality review
  • JSON-RPC
  • gRPC
  • WebSockets
  • REST
  • observability and tracing for AI systems
  • distributed systems
  • privacy-aware telemetry
  • data retention
  • secure client/service communication
  • enterprise compliance requirements
  • sandboxing
  • safety boundaries
  • permissions
  • policy enforcement for AI-initiated actions
  • modernizing legacy C++ systems
  • improving developer productivity
  • creative tools
  • productivity applications
  • developer tools
  • enterprise software
  • complex desktop applications

What the JD emphasized

  • deep C++ expertise
  • strong C++ engineer first
  • significant depth in C++
  • Expertise in modern C++ design, memory management, concurrency, API design, debugging, and systems-level performance.
  • Hands-on experience integrating AI, LLMs, agents, or model-backed systems into production or production-adjacent software.
  • Practical understanding of AI system failure modes, including hallucination, tool-calling errors, incomplete context, multi-turn drift, nondeterminism, and unreliable outputs.
  • Experience designing reliable interfaces between AI systems and deterministic software systems.
  • Strong systems architecture skills, including tradeoff analysis, technical strategy, and architecture documentation.
  • Experience with testing at scale, including unit tests, integration tests, CI validation, regression testing, and quality gates.
  • Ability to lead through influence across engineering, product, design, platform, security, and data science teams.

Other signals

  • design and build native infrastructure
  • AI-powered features to operate safely, predictably, and efficiently
  • practical AI fluency
  • integrate them into production workflows
  • design systems that make AI useful, reliable, observable, and secure
  • build evaluation frameworks for AI workflows
  • Define guardrails for AI-driven actions
  • Create privacy-conscious tracing, observability, and diagnostics for model-backed systems
  • Partner with product, data science, security, legal, and AI governance teams to ensure AI capabilities meet quality, safety, and compliance expectations