One backend, many apps

Unpod is infrastructure, so the products built on it differ: a contact center, a personal assistant, an email-first agent. Underneath, every one of them uses the same six layers.

Your application

ActionsYour APIs · webhooks · Sheets

AgentsHarness · tools · flows · human handoff

Context EnginePeople · history · knowledge · outcomes

ModelsSTT · TTS · LLM · failover

ChannelsCall · chat · WhatsApp · SMS · email

TelephonyNumbers · SIP · PSTN · media

The gallery

Unpod Platform

Our own product - Studio, Spaces, calls and analytics - built on this backend. App number one.

Contact center

Runnable today. One prompt, fifty calls, one report - with a handoff to a human when it matters.

Personal assistant

Illustrative. The agent is the interface, and it remembers the user from last time.

Email-first agent

Illustrative. An agent that lives in the inbox and calls when a conversation needs a voice.

Runnable means every step on the page works today. Illustrative means the page shows the design; it lists what the app waits on and what you can build today instead.

Which layers each app uses

Layer	Contact center	Personal assistant	Email-first agent
Telephony	Yes	Yes	For the call step
Channels	Voice	Voice, chat, WhatsApp*	Email*, voice
Models	Yes	Yes	Yes
Context Engine	Per call	Across channels*	Across channels*
Agents	Yes	Yes	Yes
Actions	Sheets, webhooks	Your APIs	Your APIs

* In build - each app page says what it waits on.

​The gallery