How to use Perplexity Computer to build a custom slack inbox (full tutorial)
Yash Tekriwal (guest), Claire Vo (host)
In this episode of How I AI, featuring Yash Tekriwal and Claire Vo, How to use Perplexity Computer to build a custom slack inbox (full tutorial) explores build a custom Slack inbox using Perplexity Computer and connectors Yash explains how 100–150 daily Slack notifications create anxiety even though most are low-priority FYIs, motivating a system to classify messages by urgency and type.
Build a custom Slack inbox using Perplexity Computer and connectors
Yash explains how 100–150 daily Slack notifications create anxiety even though most are low-priority FYIs, motivating a system to classify messages by urgency and type.
He first prototypes a deterministic Slack “digest” pipeline using Slack APIs and an AI coding agent (OpenClaw) while reserving AI primarily for the subjective step of message categorization.
He then uses Perplexity Computer to rapidly turn the digest into a usable web UI—a three-column Kanban board with filters and an “Archive All” action that clears FYIs from both the dashboard and Slack.
The conversation argues that AI enables a long-tail “micro-software” ecosystem: users can build highly specific workflow extensions to products like Slack, and small paid tools can thrive without venture scale.
Additional Perplexity Computer use cases include automating meeting follow-ups across Notion and Asana and quickly prototyping persona-based learning journeys for Clay University by leveraging browser-based visual context.
Key Takeaways
Separate “message retrieval” from “message judgment.”
Yash uses deterministic Slack API logic (timestamps, thread context, mention types) to fetch the right items, and applies AI only where human-like judgment is needed (action required vs read vs FYI).
Get the full analysis with uListen AI
A digest is helpful, but a UI changes behavior.
The initial text digest reduced noise but was still draining to scroll; the Kanban dashboard made prioritization, navigation, and bulk actions (like archiving FYIs) fast enough to become a daily habit.
Get the full analysis with uListen AI
Perplexity Computer’s advantage is orchestration, not just a model.
Yash highlights multi-model “ensemble” execution, parallel tasks, and automatic troubleshooting loops that reduce the repetitive re-prompting common in single-model coding workflows.
Get the full analysis with uListen AI
Cloud-native agents make connectors and auth feel “sticky.”
Because Perplexity Computer runs in the cloud and shares connector authentication across apps and deployed UIs, it avoids repeated token setup and can often re-authenticate or proceed via browser actions.
Get the full analysis with uListen AI
Use connectors to close the loop from insight to action.
Instead of only summarizing meeting transcripts, Yash describes extracting action items from Notion notes, routing them into Asana, and drafting messages/emails—turning AI from reporting into execution.
Get the full analysis with uListen AI
AI enables “micro-software” that lives on top of incumbent SaaS.
Rather than replacing Slack, the approach adds a personalized layer that matches an individual’s mental model—supporting the idea that many small paid extensions can be sustainable without massive TAM.
Get the full analysis with uListen AI
Maintain skills/tools iteratively where models are weak.
When an agent repeatedly fails (e. ...
Get the full analysis with uListen AI
Notable Quotes
“I truly wake up to maybe 100 to 150 new Slack notifications… 60 to 80% are more in the FYI category.”
— Yash Tekriwal
“You can use AI to do a task for you… or you can use AI just to build a tool that would've been much harder to build before.”
— Claire Vo
“I can just go ahead and click this Archive All button… and then those notifications will also disappear on my Slack.”
— Yash Tekriwal
“My dream is for someone else to watch this video… and then I can go pay that person $15 a month for this app… because I would happily pay that.”
— Yash Tekriwal
“There has always been this long tail queue of… niche customer requests… And the answer as a reasonable SaaS PM is like, ‘Never.’”
— Claire Vo
Questions Answered in This Episode
In your Slack digest logic, which specific Slack timestamps and API endpoints were most critical for deciding what counts as “unread” versus “seen in-thread”?
Yash explains how 100–150 daily Slack notifications create anxiety even though most are low-priority FYIs, motivating a system to classify messages by urgency and type.
Get the full analysis with uListen AI
What rubric or prompt did you use to classify messages into “action required,” “need to read,” and “FYI,” and how did you validate that the model was being consistent?
He first prototypes a deterministic Slack “digest” pipeline using Slack APIs and an AI coding agent (OpenClaw) while reserving AI primarily for the subjective step of message categorization.
Get the full analysis with uListen AI
How does the “Archive All” button technically map to Slack actions—mark as read, archive channel, dismiss mention, or something else—and what are the edge cases?
He then uses Perplexity Computer to rapidly turn the digest into a usable web UI—a three-column Kanban board with filters and an “Archive All” action that clears FYIs from both the dashboard and Slack.
Get the full analysis with uListen AI
You mentioned Perplexity Computer uses multiple models (Sonnet/Gemini/Opus) across steps—how do you notice when model-switching helps versus hurts?
The conversation argues that AI enables a long-tail “micro-software” ecosystem: users can build highly specific workflow extensions to products like Slack, and small paid tools can thrive without venture scale.
Get the full analysis with uListen AI
What security or privacy considerations should teams evaluate before connecting Slack/Gmail/Notion/Asana into a cloud agent that can deploy shareable apps?
Additional Perplexity Computer use cases include automating meeting follow-ups across Notion and Asana and quickly prototyping persona-based learning journeys for Clay University by leveraging browser-based visual context.
Get the full analysis with uListen AI
Transcript Preview
I truly wake up to maybe 100 to 150 new Slack notifications, not even just like, "Oh, these are unread channels." Truly someone has tagged me. 60 to 80% are more in the FYI category. So my 100 to 150 that's giving me anxiety is actually more like 30 to 40 that I really need to be on top of.
You can use AI to do a task for you, like categorize things, summarize things, or you can use AI just to build a tool that would've been much harder to build before with very straightforward APIs and structured data.
Exactly. Think about like a Kanban style board. You have in red on the left action required, urgent, Yash needs to get back to it. In the middle, we've got a yellow need to read column, and then on the right in green, much more easy, I have a bunch of FYIs. I can just go ahead and click this Archive All button. They'll disappear from the dash, and then those notifications will also disappear on my Slack.
Ugh, that's magic. And this is such a better way to just get through your queue.
My dream is for someone else to watch this video and say, "I wanna build that app on top of Slack," and then I can go pay that person $15 a month for this app to be maintained and used, and then I can file bug reports with them instead of having to fix it myself. 'Cause I would happily pay that.
[upbeat music] Welcome back to How I AI. I'm Claire Vo, product leader and AI obsessive here on a mission to help you build better with these new tools. Today we have Yash Tekrawal, head of education at Clay, and he is a hyper-optimizer showing us how he uses Perplexity Computer to work through the hundreds of Slack messages he gets every day. We're also gonna debate is SaaS really dead? Let's get to it. This episode is brought to you by Guru, the AI layer of truth for your company's knowledge. Here's the problem. Your AI is only as good as the information you feed it. Most companies are getting confident but wrong answers from AI because their underlying knowledge is outdated, incomplete, or just plain incorrect. Bad information doesn't just slow you down. It costs you money and puts you at risk. Guru solves this by adding a verification layer between your company's knowledge and AI tools. Instead of just hoping your AI gets it right, Guru automatically scores content for accuracy, flags outdated information, and ensures your team gets trustworthy answers every time. It works with the tools you already use so you don't have to change how you work. Thousands of companies trust Guru to keep their AI accurate and compliant. Ready to stop playing Russian roulette with your company's knowledge? Visit getguru.com to learn more. Welcome to How I AI. Yash, I'm so excited. We've been trying to make this happen for months, and we've been trying to make it happen over Slack for months. And what I love about that is what we're gonna start this episode off with is how you get yourself unburied from the deluge of Slack messages and emails and work you have to do on a daily basis.
Install uListen to search the full transcript and get AI-powered insights
Get Full TranscriptGet more from every podcast
AI summaries, searchable transcripts, and fact-checking. Free forever.
Add to Chrome