FeatherNet.store

Visit FeatherNet.store →

An end-to-end AI bird platform — smart feeders with cameras, live AI video streaming, real-time bird identification, a community social network, and a custom commerce storefront. Built top to bottom.

What it is

FeatherNet is a full backyard-birding platform — not just a store. Smart bird feeders stream live video to the cloud, an AI vision model identifies the species in real time, members watch their own and other people’s feeder cams, and a social layer turns it into a community: follows, comments, leaderboards, rare-sighting alerts, and shareable clips. The storefront sells the hardware and accessories that feed the whole loop.

What we built

• Smart-feeder firmware integration — feeders wake on motion, push compressed video to the streaming pipeline, and report telemetry back to the platform.
• Live AI video streaming — low-latency RTMP/HLS pipeline so subscribers watch their feeder (and follow other backyard cams) in near-real time, on mobile and web.
• AI bird identification — vision model running against the live frames, tagging species with confidence scores and saving highlight clips automatically.
• Bird-watcher social network — profiles, follows, comment threads on sightings, weekly species leaderboards, rare-bird alerts to your feed, and clip sharing.
• Custom commerce storefront — the front door at feathernet.store, with the parallax leaf-particle forest, light/dark theming, and a product catalog tied directly to the platform’s hardware and subscriptions.
• Mobile-first design system — every surface, from the cam viewer to the checkout, ships with a unified token system that re-themes the whole platform at the toggle of a switch.

The complexity

• Real-time video at consumer-internet uplink speeds — encoding, edge buffering, and AI inference all running on a budget that has to make sense per feeder.
• Identification accuracy that holds up across weather, lighting, and partial-view shots — with confidence-aware UX so a “maybe Cardinal” still feels useful.
• A social graph that scales without becoming spammy — feeds curated by interests, location, and rarity rather than chronological flood.
• Animated forest particle field running smoothly on mobile alongside a real product catalog and checkout flow.
• One design system spanning hardware-focused UI (live cam, alerts) and traditional commerce (PDPs, cart, account) — rare to see done coherently.

Outcomes

• A platform that feels like one product, not a store bolted onto a service.
• Live AI streaming + ID running at price points that work for a consumer audience.
• A social loop that gives every feeder owner a reason to come back daily.
• Storefront, app, and admin built from one codebase and one design language — fast to ship new features across all surfaces.

Where Think New helped

Think New designed and built FeatherNet end-to-end — the streaming pipeline, the AI identification model integration, the social network, the storefront, and the design system that holds it all together. From “concept on a whiteboard” to live platform.