Introducing Automations in Perplexity Computer. Automations are for ongoing work. They can take action in response to event-based triggers or on a schedule. Automations work with your memory, skills, and connected apps like Slack, Gmail, Outlook, and Linear.

Read more about Automations in Perplexity Computer:

I don’t know anyone that still uses perplexity

Yeah don't use perplexity - they straight up scam you and don't honour trial cancels, charge your card, then tell you it's your fault. Joke of a company.

This is a huge update. Running tasks on a schedule will save so much time.

wait so a slack event and a schedule can live in the same automation?

Whoa this is useful move

the real unlock is not scheduling tasks it is giving them enough memory and context to make the result dependable

This is where AI agents get really interesting. Automations + memory + connected apps is a serious combo.

The automation angle actually makes sense here — it's the first time I've seen them frame AI as something that handles ongoing work instead of just one-off tasks.

This is a strong step toward agents that actually stay useful over time. Event triggers plus memory and connected apps is the difference between a one-off assistant and something you can trust with ongoing work. Looking forward to seeing how the review flow holds up in real use.

定时跑周报省心，事件触发适合盯告警，就是Slack授权偶尔掉

Perplexity automations is nothing more than a joke. You try to build one. Less than 25% done & it has eaten whole month’s $200 worth of 10000 credits. You spend another $600 to buy enough credits to finish the task. And you are delivered half baked product.

Does perplexity stand anywhere among Astra, Fable, or SuperGrok? I don't think so.

트리거와 메모리가 붙을수록 반복 작업보다 권한 경계가 중요해집니다. Slack·Gmail 이벤트는 초안 생성과 외부 전송을 분리해 검증하는 흐름이 현실적이겠네요.

Event triggers plus a schedule is the combo that turns a one-off computer into an actual workflow.

Event triggers plus inbox access is where this gets useful and where it gets scary. Are there per-run action or spend limits? That's the first thing any ops lead will ask.

okay but how many actually run without me babysitting them

Automations on Perplexity Computer with Slack/Gmail/Linear triggers is the step from answer engine to ongoing work. Event-based, not just scheduled, is the part I will try.

Oh hello... you're back It's been a while my old friend

connected apps part is what makes this way more useful

finally an ai that can ruin my life at 3am instead of just noon

Linking Linear tickets to automated updates can shrink project lag profoundly, enhancing temporal efficiency in workflows.

Scheduled and event-based automations turn Perplexity from a research tool into something that can manage ongoing work. Would someone from Perplexity come on my show to discuss Computer and how you keep these agents under control?

event-triggered automations are the ones that actually save time for me. schedule-only ones still miss stuff for hours. once i wired a real event trigger instead of polling on a timer, half my daily-check habits went from next-morning to instant.

Event-based triggers is the part that turns this from a tool into a system. It's the same jump email went through: a newsletter you remember to send vs. a flow that fires when someone abandons a cart. The scheduled stuff saves time, the triggered stuff makes money.

Is someone still running subscriptions on this company's products? 😂😂

Perplexity Computer moving from one-off tasks to scheduled Slack, Gmail, and Linear work is where "AI assistant" starts feeling less like a chat window and more like a junior ops person who never sleeps

Always-on automations are useful. The part operators still under-spec: what event counts as done, and who can revoke the schedule without opening a ticket.

This is a big step toward truly hands-off workflows—scheduled and event-triggered AI actions connected directly to the tools you already use.

定时自动跑起来 比我自己记得住靠谱

The category is converging fast: memory + triggers + connected tools + persistent execution. The hard part now is less “can the agent act?” and more “can I trust it to keep acting correctly when I’m not watching?”

you seem to be 6 months late