Service Supply Chain AI Engineer

KLA; Ann Arbor (MI); On-site; USD 140,000 - 200,000; Full time

Job summary

KLA in Ann Arbor is seeking a skilled AI/ML engineer to build and deploy predictive models and internal tools for spares planning. You will design graph-based representations of spares, parts, tools, and configurations, and deliver self-serve analytics that influencers across planning teams rely on.

You will collaborate with SMEs to validate model behavior, implement robust data pipelines, and ensure tools are usable and explainable in an enterprise setting.

Qualifications

• Strong Python skills for data and ML development.
• Experience building customer demand prediction models or operational decision problems.

Responsibilities

• Build AI tools that planners actually use to operationalize analytics for spares planning decisions.
• Design and implement graph databases to represent spares demand, parts, tools, configurations and signals.
• Develop predictive ML models for demand forecasting and planning signals.
• Productionize tools with testing, monitoring, and documentation for enterprise use.

Skills

Python

ML/AI development

Graph databases

SQL

Data pipelines

Stakeholder collaboration

Tools

APIs

Dashboards

Job description

To make electronics, you need chips, wafers, transistors, reticles, and... To make these, you must see, test and manufacture them at scale-faster and better than ever before. That's where KLA comes in. Whether you're early in your career or an experienced professional, you'll solve complex challenges, work alongside brilliant minds and help shape the future of technology.

Group/Division

The KLA Services team consists of Service Sales, Marketing, Spares Supply Chain Management, Field Operations, Engineering, Product Training, Digital Solutions and Analytics, and Technical Product Support. Our services organization maximizes the value of our customers' KLA assets - with highly trained service and product support engineers providing installation services and 24/7 technical support and parts delivery through our extensive supply chain network.

What You'll Do

In this role, you will play a key part in advancing business priorities by delivering high-impact work across your area of expertise.

1. Build AI tools that planners actually use
   • Develop and maintain internal tools (apps, dashboards, workflows) that operationalize advanced analytics for spares planning decision-making.
   • Translate planning problems into well-scoped product requirements: user journeys, success metrics, data needs, and rollout plans.
   • Create "self-serve" tools that reduce manual effort and scales insights across the organization
2. Design and implement graph databases for AI use cases
   • Define the graph data model (nodes/edges, ontology/taxonomy, temporal relationships, metadata) to represent spares demand, parts, tools, configurations, sites, and operational signals.
   • Build ingestion pipelines and data quality checks to keep the graph accurate, explainable, and trusted.
   • Enable AI and analytics on top of the graph: graph traversals, similarity search, embeddings, and graph ML patterns that support decision tools.
3. Apply ML / AI to spares planning outcomes
   • Develop predictive models for demand forecasting (including intermittent/long-tail behavior), demand drivers, and related planning signals.
   • Support inventory planning improvements (e.g., safety stock, multi-echelon thinking, service-level tradeoffs) by connecting model outputs to actionable recommendations.
   • Partner with SMEs to validate model behavior, define guardrails, and ensure outputs are usable and explainable in operational settings.
4. Productionize: reliability, governance, and MLOps
   • Implement testing, monitoring, documentation, versioning, and performance practices so tools are robust and maintainable.
   • Establish repeatable deployment patterns (dev/test/prod), model monitoring, and data lineage appropriate for enterprise planning environments.
   • Create clear documentation and enablement materials so tools can be adopted broadly (not just by technical users).

What Success Looks Like (examples of outcomes)

• Planners can answer critical questions faster (and with less manual wrangling) because data and relationships are captured in a reusable graph and exposed through intuitive tools.
• Improved forecast quality and earlier detection of demand changes for targeted segments (especially long-tail/intermittent parts), leading to fewer expedites and fewer stockouts.
• Reduced avoidable inventory buffers through better segmentation, variability modeling, and decision support tied directly to planning actions.

Minimum Qualifications

Skills & Experience Needed

Required skills (must-have)

• AI / ML & analytics
  • Strong Python skills for data and ML development (pandas/numpy, ML libraries, model evaluation).
  • Experience developing customer demand prediction models, or other operational decision problems.
• Graph theory & graph data
  • Solid foundation in graph theory concepts (graph modeling, connectivity, centrality, communities, bipartite/multipartite graphs, temporal graphs).
  • Hands‑up? Use context.
• Software & data engineering
  • Strong SQL and data modeling; ability to build reliable pipelines across large enterprise datasets.
  • Experience building production services or internal tools (APIs, web apps, dashboards) with a focus on usability and maintainability.
• Collaboration
  • Proven ability to work with non-technical stakeholders, convert ambiguous business needs into effective tools, and drive adoption/change management.

Preferred skills (nice-to-have)

• Supply chain planning experience (service parts, inventory optimization, safety stock, service-level tradeoffs, replenishment/network concepts).
• Experience with probabilistic forecasting approaches and intermittent-demand methods.
• Knowledge graphs / ontology design, entity resolution, and "semantic" modeling patterns.
• Experience integrating LLMs with structured data (RAG patterns, tool calling, natural-language-to-query workflow