Can AI replace Amplitude?

Build a closest honest personal substitute for Amplitude in an empty repository.
Use Next.js 15, TypeScript, ClickHouse, PostgreSQL, and Docker Compose; do not offer alternative stacks.
The core loop is: collect privacy-conscious first-party events, answer a small set of product questions with funnels, cohorts, and retention views, and retain raw data under the owner's control.
Make the first run work locally with one documented command.
Store all user data locally by default and make export straightforward.
Put secrets in .env, ship .env.example, and never commit credentials.
Ship a lightweight browser SDK for page views and explicit custom events.
Create projects, API keys, environments, event names, and a documented event schema.
Ingest idempotently, filter obvious bots, truncate IP data, and avoid fingerprinting.
Build dashboards for visitors, sessions, funnels, retention, sources, pages, and events.
Expose filters by date, environment, device, country, referrer, and selected properties.
Add retention controls, raw-event export, deletion, health checks, and backup instructions.
Include clear empty, loading, success, and recoverable error states.
Add input validation, safe filenames, and graceful handling of unavailable APIs.
Write focused tests for the core transformation and one end-to-end happy path.
Create a README with setup, architecture, permissions, data location, and backup steps.
Do not add accounts, billing, telemetry, analytics, or a hosted control plane.
Do not claim to reproduce proprietary data, network liquidity, regulated access, or frontier infrastructure.
Deliberately leave out cross-site identity graphs.
Deliberately leave out session replay and automatic DOM capture.
Deliberately leave out warehouse-scale reverse ETL and enterprise governance.
Finish by running the tests and listing the exact commands used.