New courseCourse
Design Thinking in AI
Take one real workflow from idea to a working AI prototype, using design thinking to decide what to build and how to check it.
Pricing and dates announced soon
Overview
What you'll learn
You learn the design thinking loop, empathize, define, ideate, prototype and test, and apply it to AI products and workflows. By the end you have taken one real workflow from idea to a working AI prototype, tested with the people who use it.
- Map a workflow with the people who do it, and mark where AI could help and where it should not
- Frame AI problems in plain language, with no product names, and score them for impact, ease, data and risk
- Match each problem to an AI pattern: drafting, summarizing, routing, lookup or classifying
- Prototype with prompts and simple agents in hours, with a human check and an off switch
- Test AI outputs with real users against pass criteria you set before you look
- Finish with a working AI prototype of one real workflow, a test card and a 90-day plan
Audience
Who it's for
A good fit for
- Owners, operators and team leads who want AI to fix a real workflow, not add another tool
- Product, ops and cross-functional teams building AI features or internal AI tools
- People who tried AI assistants and saw nothing change in how work gets done
- Facilitators who want AI projects to end with an owner and a test plan
Not the right fit for
- Anyone looking for a data science or model training course
- People hunting for a ranked list of chatbots
- Anyone who wants customer-facing AI with no human review
Outline
Course outline
8 modules. Each one ends with a short exercise you complete on your own work.
- Module 01
Design thinking for AI
The five stages, empathize, define, ideate, prototype and test, applied to AI products and workflows. Why AI projects stall when they start with the tool: design the work, then pick the tool.
- Module 02
Empathize: map the real work
Interview the people who do the workflow. Map steps, handoffs, hours and pain, and mark the parts that must stay human.
- Module 03
Define: problem framing for AI
Write problem statements as who, friction, situation and cost. Score impact, ease, data readiness and risk, and decide whether it is an AI problem at all.
- Module 04
Ideate: patterns and constrained options
Match problems to five AI patterns. Generate ten ideas inside real limits on time, budget and skill, then pick one primary and one backup.
- Module 05
Prototype: prompts and agents
Build prompt prototypes, a voice card for drafting, an assistant grounded in your own documents and a simple agent or automation with a named owner and a kill switch.
- Module 06
Test: evaluating AI outputs with users
Write a hypothesis and pass criteria first. Put real outputs in front of real users, log errors and decide to kill, tweak or keep.
- Module 07
Responsible AI and risk
Human review on anything customer-facing, no invented facts, disclosure when it matters, private data kept out of the wrong tools and an owner for every automation. A short risk check for each prototype.
- Module 08
Capstone: idea to working prototype
Take one real workflow through all five stages. Finish with a working AI prototype, a test card with results and a 90-day plan to harden, kill or expand.
Format
How it works
- How it runs
- Self-paced. Short video lessons you can take in any order that suits your week.
- Where
- Online, on any device.
- Materials
- Workflow map template, AI problem statement cards, opportunity scorecard, prototype brief, AI output test card, risk checklist and a 90-day plan template.
- Tools
- Built around the AI assistants and no-code tools you already have. The core path needs no coding.
- Pricing and dates
- Announced soon. Join the waitlist and we will email you first.
FAQ
Questions
When does it start, and what does it cost?
Pricing and dates will be announced soon. Join the waitlist and we will email you first.
Do I need a technical background?
No. The course uses everyday AI assistants and no-code tools. If you can map a process on paper and write a clear email, you can do the capstone.
Which AI tools will we use?
The method is tool-agnostic. Lessons show examples with common AI assistants, and you build with whatever you already have access to. Tools change; the method does not.
What do I need for the capstone?
One real, repetitive workflow from your own work and, ideally, one or two people who do it and can try the prototype. Use sample or anonymized data, not private customer records.
Can my team take this together?
Yes. Obra Workshops runs live sessions on the same topics for teams. Email us and we will point you to the right option.
Design Thinking in AI
Join the waitlist.
Add your name and we will email you when pricing and dates are announced.
Pricing and dates announced soon. Questions? info@obralearning.com
Keep exploring