I’m ZhiChao. I lead product decisions, organize AI-assisted implementation, and use evaluation and review to turn ideas into work people can inspect.
Computer Science student Preparing for AI / computing graduate study Open to engineering & product opportunities
PORTFOLIO / 2026.09Three complementary projects, with the decisions, working examples, and evidence behind them.
An editable go-to-market workspace that connects a founder’s assumptions, recorded evidence, decisions, and next actions.
Explore the case
A practical guide and small Python toolkit for turning evaluation requirements into explicit criteria, tolerance checks, and reviewable task packages.
Explore the caseA Windows dispatcher that turns installed AI CLIs into named workers with explicit routing, capability checks, and structured execution results.
Explore the caseThree ways to make AI work worth inspecting.
Evaluation workspaces, financial documents, reliability studies, and earlier product explorations.
Three reviewable research cases: budgeted retrieval selection, LoRA input correction and transfer stress testing, and controlled battery-capacity reconstruction, supported by evaluation tooling and personal practice.
A guided workspace for comparing model screenshots and deliverables, then revising and sharing a structured evaluation report.
An exploratory comparison of plain-text and layout-aware parsing for questions over financial tables.
A bounded audit of battery-capacity reconstruction that separates learned constraints from EMA and running-min post-processing in held-out-cell tests.
A market-research workspace with specialist analysis, visible progress, structured reports, and a handoff into LaunchLens AI.
A local Codex-to-ZCode relay experiment for bounded worker tasks, status collection, and review before integration.
A financial-analysis prototype that separates supervision, document research, and computation in a graph-based workflow.
A private prototype exploring AI support for long-term personal progress.
A private financial-document prototype exploring visual retrieval, region cropping, and agent-assisted verification.
A private research prototype for mandibular nerve segmentation in 3D CBCT images.
A private campus marketplace and services prototype for student workflows.
3 accessible demo entry points. Each case distinguishes example data, product functionality, and the scope of validation.
I study Computer Science at Guangling College, Yangzhou University. I’m preparing for overseas graduate study in AI and computing, while exploring opportunities to contribute to engineering and product work.
My work starts with clarifying the problem and organizing the tools to address it. I use AI in implementation, then question the choices, inspect the results, and guide the next iteration. Each case makes my role and the limits explicit.