On 18 August 2026, the Peiron journal recorded the completion of Phase 0, Definition and Design. Nine batches established the product charter, experience, search and perspective model, privacy and economics, iPhone and Safari intent, architecture, security and quality contracts, development process, and final definition freeze.
The freeze clarified several decisions that shape the planned first release. Peiron would remain a conventional search engine first, add source-grounded perspective discovery only where it helps, and keep AI subordinate to the user's access to sources. The first release would use no Peiron account, keep optional history on the device, and separate the platform-independent search service from its native iPhone and Safari clients.
The architecture was fixed at the level of technology families and boundaries: a Python and FastAPI modular monolith, PostgreSQL for durable entitlement and usage state, Valkey for ephemeral controls, a native SwiftUI client, and replaceable search and model providers. Exact vendors, versions, pricing, hosting, retention periods, and Safari routing were intentionally left for measurement or capability testing in later phases.
This milestone did not produce working software. The closeout states that implementation had not started and identifies a reproducible engineering foundation as the next milestone. Phase 0 made the idea coherent enough to build, but it did not validate the search quality, user experience, cost, security, or platform behavior of the future product.
← Back to Peiron