01Discover
Client discovery & FDE practice
Turn an ambiguous AI request into a scoped project with a workflow owner, a baseline, and evidence of value.
Forward deployed AI engineering / Databricks
From a client problem
to a system that works.
Learn the full delivery cycle: discover the need, engineer the data, build useful AI, and operate it with confidence.
Your field project
A governed support copilot.
01
Understand the workDiscovery, stakeholders, measurable outcomes
02
Build with evidenceReliable data, retrieval, bounded tools
03
Make it dependableEvaluation, permissions, approval, releases
04
Deliver and operateMonitoring, runbooks, client handoff
The curriculum
Learn the job, end to end.
Follow the sequence or focus on a gap. Every module includes a worked scenario, a practical assignment, and a knowledge check.
20 modules
02Build
Production Python & APIs
Build small services with explicit contracts, predictable failures, and tests that reveal real defects.
03Build
SQL & data contracts
Make business calculations reliable by defining grain, validating joins, and treating data assumptions as contracts.
04Build
Cloud & Databricks foundations
Understand storage, compute, identity, and catalog boundaries before provisioning an AI application.
05Build
Spark & Delta engineering
Build repeatable transformations and understand the execution decisions behind reliable incremental pipelines.
06Build
Orchestration & streaming
Coordinate data work, handle late events, and recover interrupted pipelines with evidence of correctness.
07Build
LLM application engineering
Choose models and prompts through task evidence, validate outputs, and design for uncertainty.
08Build
Retrieval-augmented generation
Connect answers to source evidence and diagnose retrieval separately from generation.
09Build
Governed Databricks search
Build an explicit authorization boundary around managed search and verify the full path to the answer.
10Build
Bounded agents & human approval
Let models propose actions while trusted software controls permissions, state transitions, and execution.
11Evaluate
MLflow evaluation & tracing
Build a repeatable evaluation loop that connects application traces, human judgment, and release decisions.
12Evaluate
Governance & AI security
Connect data access, model governance, threat modeling, and adversarial testing into one defensible boundary.
13Ship
CI/CD & LLMOps
Version the whole AI application and release only when the relevant engineering and evaluation gates pass.
14Ship
Enterprise integrations
Connect AI workflows to operational systems without losing authorization, consistency, or failure visibility.
15Discover
Domain prototyping
Build a small useful prototype, gather realistic feedback, and decide what must change before a pilot.
16Evaluate
Quality, latency & cost
Measure the complete task and compare architecture choices under declared, repeatable conditions.
17Ship
Deployment & platform engineering
Release a reproducible application with deliberate identity, networking, configuration, and recovery choices.
18Operate
Operations & incident response
Detect degraded behavior, recover deliberately, and communicate clearly when an AI application fails.
19Operate
Client delivery & pod leadership
Keep technical work aligned with decisions, dependencies, scope, and the people who will operate it.
20Operate
Portfolio & interview practice
Present verifiable evidence of your work and explain your decisions under unfamiliar constraints.
No modules match this search. Try a broader topic or choose All modules.
Weeks 21–24 / Integrated capstone
Show the whole system.
Defend every decision.
Build a support copilot with authorized evidence, human-approved actions, repeatable evaluation, and an operational handoff.
Explore the capstone- A client discovery brief and measurable baseline
- A tested application with explicit access boundaries
- An evaluation report, release plan, and runbook
- Executive and engineering demonstrations