Thiên Thành
VIEN

Speaking

I use working demos to show what AI can do, and where it must stop.

I do not want an audience to remember only the name of a tool. I want them to see a real workflow, where the decision remains and which evidence should be checked before accepting a result.

Real workflowsWorking demosExplicit boundaries

Speaker poster for Thiên Thành at Earn in Public Summit Q3/2026

Latest talk · 17/10/2026

AI helps me build faster, but it does not get to decide

I show how a request moves through Intake, where AI can read and propose but must stop before human approval, action and acceptance.

The talk starts with a situation I encountered while building Intake: requests arrive from many places, and AI can structure them quickly, but speed cannot replace approval, provenance or a definition of done.

In the demo, AI captures and proposes. People still approve, act and inspect the evidence before work is considered complete.

Earn in Public Summit Q3/2026 · Talk and live demo

View the talk and demo

What I want people to take away

Tools will change. Delegating work well is the durable skill.

You understand the context and responsibility of your work better than any AI model. The important skill is knowing what to delegate, which authority to grant and what evidence to require before accepting the result.

01

Choose the right work

Separate tasks AI can read, structure and propose from decisions people need to retain.

02

Bound the authority

Design stop points, responsible reviewers and the scope a system is allowed to change.

03

Inspect the evidence

Do not confuse a fluent answer with a correct result; define the source, state and completion criteria.

Speaking invitation

I speak about systems I have directly built and operated.

I am a good fit for sessions that examine one specific workflow with founders, product, operations or engineering teams. The format can be a short talk, workshop or live demonstration, depending on the audience and the question we need to clarify.

I will state what is running, what remains experimental and the limits of what I demonstrate. Customer data, internal architecture and private information do not appear in public demos.

If you would like to connect, send the topic, the audience and what you want them to be able to do after the session. I will respond and participate when the subject is a good fit and can bring practical value to the community. Thank you for your interest.

Email Thiên ThànhWhatsApp