Primary project
Plato
An AI-assisted product workbench for turning ambiguous product work into durable plans, documents, and verifiable execution loops.
138k+ lines of code 241k+ with docs long-running iteration
Read the case study Projects
Project proof first, focused on Plato.
Plato is the strongest project evidence on this site, showing product judgment, Agent architecture, media demos, and long-running iteration.
Portfolio / Resume link
Zhang HaoAI Product Builder
I build AI-assisted product systems with a focus on durable workflows, verifiable delivery, and clear product judgment.
Articles
Selected writing
Writing shows the product retrospectives, architecture research, and Agent product thinking behind the project evidence.
Jun 11, 2026
Failure Lessons From One Month Building An Agent Product With AI
After a month of AI-assisted Agent product development, the real problems were not code generation but product loops, state models, recoverability, workflow gates, and demo paths.
Jun 11, 2026
Do Not Start Agent Products With Screens: State Models Matter More
Agent product pages are not static information displays. Task semantics, state models, API contracts, event reducers, and recoverability should come before UI.
Jun 09, 2026
Product Boundaries Of Multi-Agent Parallelism
Default multi-agent parallelism is losing its appeal. The valuable direction is controlled parallelism built on task isolation, context governance, permission control, and result verification.
Contact
Open to AI product and Agent product manager conversations.
For a deeper review, start with the resume, Plato case study, GitHub profile, or direct contact entry.