Siemens Digital Industries Software is a leading provider of solutions for the design, simulation, and manufacture of products across many different industries. Formula 1 cars, skyscrapers, ships, space exploration vehicles, and many of the objects we see in our daily lives are being conceived and manufactured using our Product Lifecycle Management (PLM) software.
Are you passionate about AI and excited about taking Generative AI capabilities from prototype to production inside a real enterprise We are looking for a dedicated AI Engineer to join the CApS AI Engineering team and help build and productionise embedded AI capabilities across CApS—delivering solutions that are secure, scalable, measurable, and operationally ready.
Role Purpose-
- Embedded AI delivery (copilots, assistants, agentic workflows, retrieval-augmented experiences),
- Platform & tooling foundations (golden paths, reusable templates, reference integrations),
- Governance-by-design (risk controls, auditability, privacy/security expectations),
- Enablement (helping other CApS teams adopt AI patterns confidently and safely).
Role Clarification (Included and Excluded Responsibilities)
- A partner in designing, building, and operating AI-enabled capabilities that solve real operational and product problems.
- An enabler of scale through reusable patterns, templates, and guidance.
This role is not intended to be:
- A helpdesk or on-demand support line for ad hoc experimentation.
- A centralized team that does AI for everyone end-to-end.
- A dumping ground for unfinished prototypes.
Key Responsibilities-
As an AI Engineer, you will design, build, and maintain AI-enabled capabilities powered by Large Language Models (LLMs), working closely with product and platform teams.
You'll make a difference by:
- Designing & delivering embedded AI capabilities (copilots, agents, workflows)
Designing and implementing AI features using validated patterns such as RAG, tool calling, agents, and multi-step workflows.
Integrating AI capabilities into existing services and applications via APIs and well-structured backend components.
Ensuring AI features are scalable, high-performance, and aligned with enterprise expectations.
- Productionising AI (from intake to operation)
Moving AI use cases through a practical lifecycle: intake → risk classification → gated approvals → build → test → deployment → operation.
Defining operational readiness (runbooks, monitoring, rollout strategies) and supporting production incidents and solving when needed.
- Quality, evaluation, and continuous improvement
Developing and delivering evaluation plans and test harnesses for AI systems (offline tests, regression checks, retrieval scoring, and reliability checks).
Defining what good looks like using clear metrics (e.g., success rate, latency, cost, failure modes) and using those metrics to drive iteration.
Implementing fallback, recovery, and human-escalation mechanisms for failure scenarios where appropriate.
- Observability and measurable operations
Instrumenting AI services for tracing, logging, cost signals, and feedback loops to improve quality and reliability over time.
Contributing to operational dashboards and alerting patterns to support debugging and production learning cycles.
- Platform & tooling (golden paths)
Contributing to reusable foundations: templates, CI/CD hooks, reference implementations, and integration patterns that make safe AI delivery repeatable.
Partnering with platform and security collaborators to embed security, access control, and compliance requirements into paved paths.
Building and integrating cloud-native AI components on major platforms (AWS and Azure), including handled agent/runtime and retrieval services (e.g., Amazon Bedrock AgentCore and Knowledge Bases; Microsoft Foundry and Foundry Tools) where appropriate for the workload and compliance context.
- Enablement & adoption across CApS
Pairing with development teams to onboard them to approved AI patterns; supporting adoption through workshops, office hours, and documentation.
Helping accelerate the organization's shift toward AI-assisted products and AI-assisted engineering through practical guidance and shared standards.
- AI-assisted engineering (policy-aligned, high-cadence)
Using Siemens-approved AI-assisted engineering tools as part of day-to-day work (coding, refactoring, debugging, spec authoring, review, and analysis).
Working in a way where AI-assisted development is the default: keeping work small, well-scoped, and documented so progress is fast and reviewable.
Following mandatory legal and compliance guidance for AI-assisted software engineering, including human verification and secure coding practices.
Skills & Qualifications –
- Bachelor's or master's degree in computer science, AI, Software Engineering, or a related technical field (or equivalent practical experience).
- Strong software engineering skills in Python, including clean code practices, testing field, and refactoring.
- Hands-on experience building LLM-enabled systems (beyond prototypes) and understanding model behavior, failure modes, and structured outputs.
- Practical experience with RAG (retrieval, chunking/context strategies, and retrieval quality evaluation).
- Practical experience with agents/tool calling and multi-step orchestration.
- Strong fundamentals in Git and modern collaboration workflows (branching strategies, merge requests, code reviews, conflict resolution) and comfort working in CI/CD-driven delivery.
- Experience working in an Agile or otherwise well-defined engineering workflow (e.g., Scrum/Kanban) with clear backlog field, incremental delivery, and strong engineering hygiene.
- Excellent documentation skills: ability to produce clear, durable engineering documentation (design notes, ADRs, runbooks, onboarding docs) as a normal part of delivery.
- Experience working with at least one major cloud platform (AWS or Azure) in a production engineering context, including deploying and operating AI-enabled services.
Preferred/Good To Have-
- Proven, day-to-day use of Claude Code (or equivalent agentic AI-assisted coding workflows) to ship faster while maintaining quality. If you are not currently using AI-assisted coding daily, you should expect a steep ramp-up and be ready to adopt Claude Code immediately as part of how the team operates.
- Hands-on experience with handled GenAI/agent and retrieval services on cloud platforms, such as: Amazon Bedrock AgentCore (agent runtime / deployment and operations at scale) Amazon Bedrock Knowledge Bases (led RAG/retrieval) Microsoft Foundry (enterprise AI platform for models/agents/tools) Foundry Tools (prebuilt AI tools/APIs used in AI applications)
- Experience with orchestration frameworks (e.g., LangChain, Semantic Kernel, PydanticAI, or similar).
- Experience with vector stores / semantic retrieval systems and embedding strategies.
- Experience with observability stacks and distributed tracing for AI services.
- Experience operating services in a production environment with performance, reliability, and security expectations.
- Experience working in supervised enterprise environments and applying privacy/security controls by design.
Why us
Working at Siemens Software means flexibility - Choosing between working at home and the office at other times is the norm here. We offer great benefits and rewards, as you'd expect from a world leader in industrial software.
A collection of over 377,000 minds building the future, one day at a time in over 200 countries. We're dedicated to equality, and we welcome applications that reflect the diversity of the communities we Work in. All employment decisions at Siemens are based on qualifications, merit, and business need. Bring your curiosity and creativity and help us shape tomorrow!
Siemens Software. Transform the Everyday
#SWSaaS
