Enterprise AI Engineering Intermediate

AI Engineering Intermediate: Building Enterprise AI Capabilities at Scale

Accelerate your engineering teams from literacy to applied prototyping. Participants will implement retrieval-augmented generation and agent patterns, then instrument behavior with tracing and metrics. We emphasize guardrails, evaluation, and deployment choices across closed APIs and open-weights—optionally in on-prem or air-gapped contexts. This course is part of Caltech’s Enterprise AI Engineering series (Foundations, Intermediate, Advanced), a sequence that organizations can target to different skill levels; learners may progress through all three over time. All three tiers belong to the same series and can be sequenced or spaced to cascade learning across roles—enabling cohorts to advance from foundational literacy to applied prototyping and, ultimately, governed multi-agent design.

  • Learners Foundational
  • Time Client definable
  • Duration 5 Days; Definable
  • Program Type Customizable Programs
  • Certificate Type Certificate
  • Format
    Any Format/Location
  • CEUS Available
  • PDUS Available
  • Program Number AIEngI-Custom
  • Fees Group Rate
  • See full course info

Accelerate your engineering teams from literacy to applied prototyping. Participants will implement retrieval-augmented generation and agent patterns, then instrument behavior with tracing and metrics. We emphasize guardrails, evaluation, and deployment choices across closed APIs and open-weights—optionally in on-prem or air-gapped contexts. This course is part of Caltech’s Enterprise AI Engineering series (Foundations, Intermediate, Advanced), a sequence that organizations can target to different skill levels; learners may progress through all three over time. All three tiers belong to the same series and can be sequenced or spaced to cascade learning across roles—enabling cohorts to advance from foundational literacy to applied prototyping and, ultimately, governed multi-agent design.

Print Page
Enterprise AI Engineering Intermediate

Program Experience

In this hands-on course, participants design and implement a RAG pipeline and a simple agent aligned to your use cases. Labs progress from ingestion/chunking and vector search to prompt-routing and tool use. Participants add observability (tracing/metrics) to compare prompt patterns and agent actions. Takeaways include reusable notebooks, evaluation checklists, and a lightweight action outline to transition into internal hardening phases if desired. 

View our instructors

Course Info

Benefits
Topics
Who Should Attend
Schedule
FAQ

Your team will learn to:

  • Implement RAG pipelines (ingestion, chunking, retrieval, grounded generation)

  • Build and evaluate agents using LangChain, LlamaIndex, and LangGraph/LangSmith

  • Instrument tracing and metrics to compare prompt patterns and surface failure modes

  • Embed governance and reliability checks early in prototyping cycles

  • Choose deployment options (closed API, open-weight, on-prem/air-gapped) based on risk and cost

  • Plan validation gates that distinguish demos from production requirements

  • RAG fundamentals: ingestion, chunking, vector search, grounded generation

  • Agent patterns: chain-of-thought, ReAct, Reflexion; tool use basics

  • Frameworks: LangChain, LlamaIndex, LangGraph, LangSmith

  • Observability: tracing, metrics, evaluation scaffolds

  • Capstone (education prototype): build an agent using enterprise context; instrument with tracing and metrics

Software, platform, data, and DevSecOps engineers and technical leads with baseline LLM knowledge who are ready to develop RAG and agent workflows, plus managers coordinating governance prior to any internal productionization.

This customizable program can be delivered remotely, on-site, or on the Caltech campus. The program is typically 40 hours but can be tailored for your organization’s needs.

Will this course deliver a production-ready agent or application? No. Teams produce and evaluate education-time prototypes only.

Can we bring internal use cases to the labs? Yes. We encourage relevant, non-confidential use cases that help your participants practice skills aligned to your context.

Which frameworks will participants use? Participants work with LLMs and common frameworks such as LangChain, LlamaIndex, and LangGraph/LangSmith, along with standard evaluation and tracing tools.

How does governance appear in Intermediate? Governance is embedded in exercises, including policy alignment, role-based access, auditability, and reliability checks.

Can this course run behind our firewall and use our internal enterprise model endpoints? Yes. We can deliver the course on-premises, in a VPC, or in air-gapped environments and connect to your enterprise-approved model endpoints (for example, Azure OpenAI or other hosted foundation models, Bedrock in a private VPC, or self-hosted/open-weight deployments). In-class access is used for research and evaluation during the course; no production integrations or changes are performed in class. Any production access, promotion, or change management follows your enterprise policies and remains outside the scope of the course.

How does this course relate to the Enterprise AI Engineering series? Intermediate is the second tier of a three-course series designed to help companies target content to different skill levels.

Can learners progress through all three tiers over time? Yes. Learners commonly begin with Foundations and progress to Intermediate and then Advanced as responsibilities and needs mature. Experienced and proficient developers may start at Intermediate or Advanced.

Who owns the artifacts created during the course? Your organization owns the artifacts produced by participants during class; Caltech retains ownership of course materials and templates.

Our Educators

Our team of educators and guides are experts in their field – engineering pioneers, applied science visionaries, Ted-Talkers, professional facilitators, pilots, problem solvers, marketing mavens, and award-winning authors – who bring academic knowledge, practical approaches, and proven solutions to their programs.

Collectively, they have decades of experience in aerospace, communications, defense, electronics, energy, government, high-tech, pharma/medical devices, and precision manufacturing. 

Photo of Harish Kashyap

Harish Kashyap

Artificial Intelligence, Machine Learning, Data Science

Picture of Mike Krause, PhD

Mike Krause, PhD

Artificial Intelligence, Machine Learning, Data Science

Instructors

Picture of Mike Frantz, PhD

Mike Frantz, PhD

Artificial Intelligence, Machine Learning, Data Science

Photo of Harish Kashyap

Harish Kashyap

Artificial Intelligence, Machine Learning, Data Science