Local discovery for people who are past the touristy layer, from a layered retrieval engine that reads intent from a plain question and is built to surface the places list sites bury.
Sidestreet is in development toward a public release.
Sidestreet is local discovery for the SF Bay Area, built for people past the touristy layer. The app leads with recommendations, learning a user through their liked, saved and visited places and proposing where to go next, with search kept for the moments a user wants a specific place. Its core concept is Area Intelligence, a synthesized answer for an area and an occasion that accounts for atmosphere, hours, transport and parking, weather and the order activities make sense in, with a full day as the main unit of a plan. Underneath sits a layered retrieval engine over places, events and experiences across 6 venue categories, and the ablation study below priced what each of its layers buys.
Teardown
16.7% of the engine's good answers would be put out of reach by restoring the strict filter, to prevent a quality loss the study could not find.