Bosco Lam · Founder and operator

I build, experiment and write.

I explore how AI, data and emerging technologies change the way businesses are built and operated. This is where I collect ideas, frameworks and lessons from building.

Explore the ideas Read my writing

THE LEARNING LOOPIN PROGRESS

Build somethingStart with a real problem

SIGNAL

Test an ideaObserve what changes

PROOF

Share the learningWrite, reflect and repeat.

SYSTEM

FOCUS

AI-native businesses and agentic workflowsData commercialization and data productsApplied AI for business outcomesBuilding and operating technology companies

What I’m exploring

AI, data and the work of building.

Questions I return to in my writing and experiments. Some grow out of operating experience; others are ideas I’m still testing.

AI-native businesses

How do companies change when AI becomes part of the operating model?

EXPLORE

Agentic workflows, human judgment and the economics of repeatable work.

Data commercialization

How does fragmented or underused data become something a customer can use and value?

EXPLORE

An eight-stage framework from data origin and rights to measurement and commercial value.

Applied AI

Where can AI improve a business outcome, and how would we know?

EXPLORE

Bounded experiments, useful feedback and evidence of what changed.

Building technology companies

What can we learn from the decisions, constraints and trade-offs of building?

EXPLORE

Founder lessons, customer problems and ideas worth testing.

How I think

AI should reorganize the work—not decorate it.

Models make capability cheaper. Durable advantage comes from proprietary data, accountability for outcomes and a direct customer relationship. I use software to strengthen those assets—not mistake the software itself for the moat.

Explore essays on Substack ↗

Explore the eight-stage data commercialization framework →

Start with the outcome

Define the job, the acceptance criteria and the business value before choosing a model.

Own the durable edge

Build around proprietary data, accountability for outcomes and the direct customer relationship.

Keep judgment human

Automate repeatable work while preserving human decisions where context and risk matter.

Close the loop

Capture what happened after every action so the workflow improves instead of resetting.

Use durable data

Run production workflows on permissioned access that will still be there when the prototype becomes a business.

About Bosco

Founder and operator, learning by building.

I co-founded AlikeAudience. My writing and published conversations explore data products, audience intelligence and attribution. That experience informs the questions I’m asking about AI and how technology businesses are built.

This site connects those ideas with their sources and my background. Substack remains my home for long-form essays; here you’ll find reference pages and the threads connecting them.

CO-FOUNDERAlikeAudience

CURRENT FOCUSData & AI workflows

EDUCATIONUC Berkeley

Read the full profile →

If our interests overlap

What are you thinking about?

If you’re building, investing in or thinking about something around AI, data or technology businesses, I’d be interested to hear about it. A shared question can be the start of a useful conversation.

Start a conversation ↗