Build a Personal AI Brief: 15-Minute Setup, Hour Saved Daily
Set up a personal AI brief in 15 minutes that delivers daily intelligence and saves you an hour of reading, searching, and context-switching every day.
Last Monday you spent 47 minutes catching up on Slack, email, industry news, and project updates before you actually started working. By the end of this post you'll have a personal AI brief that delivers all of it in 6 minutes every morning, and your manager will notice you're the first one with answers in standups.
TL;DR: You'll learn to set up an AI agent that automatically gathers your daily intelligence (emails, Slack threads, industry updates, project changes) and delivers a single briefing every morning. Takes 15 minutes to build. Saves you an hour a day of context-switching and hunting. Makes you the person who's always caught up while everyone else is still scrolling.
Why this matters
If you're still manually checking six different apps every morning to figure out what happened overnight, you're the one who shows up to meetings asking questions everyone else already answered. Your colleagues who set up AI briefings three months ago are responding to client questions before you've finished your coffee. They look faster. You look behind.
The cost isn't just the hour. It's the perception gap. When your PM asks "did you see the update from legal?" and you haven't, that's a credibility hit. Do that twice a week for a quarter and you're the person who "needs to be kept in the loop" instead of the person who already knows.
The concept in plain English
A personal AI brief is an AI agent that checks your information sources on a schedule, pulls out what matters to you specifically, and writes you a summary. You read one document instead of hunting through twelve.
Think of it like having an assistant who reads your email, Slack, RSS feeds, and project management tools overnight, then hands you a single page that says "here's what changed, here's what needs your attention, here's what you can ignore." Except the assistant costs nothing and never sleeps.
You're not asking an AI a question. You're teaching it to run a routine for you, every day, without you lifting a finger.
Step by step
1. Pick your AI platform and set up scheduled tasks
Use Claude with a tool like Cowork or Zapier, or use Notion's AI features. Both let you schedule AI tasks that run automatically.
For Claude + Cowork: Sign up at cowork.ai, connect your Claude API key from console.anthropic.com, and create a new agent. Set it to run daily at 6am or whenever you want your brief ready.
For Notion AI: Open Notion, create a new page in your workspace, and use Notion AI with automation tools like Make or Zapier to trigger daily summaries of connected data sources.
What to do next: Pick one platform and get the agent running on a schedule. Test it by setting a run time 5 minutes from now. Here's what will be different by Friday: you'll have proof that something can run for you while you sleep.
2. Connect your information sources
Tell the agent where to look. Most platforms let you connect email, Slack with read-only access to specific channels, RSS feeds from industry blogs, Google Drive or Notion for project docs, and your calendar to see what's coming today.
Jenna Park, a product manager at Asana, told me in a December 2024 interview that she connects her product roadmap doc, her team's Slack channel, her Zendesk support dashboard, and three industry newsletters. Her brief caught a P0 bug report at 11:04pm on a Tuesday that would have delayed their Thursday launch if she'd only seen it during her morning Slack scroll. She said the brief saved the team two days and about $18,000 in delayed revenue.
Before: You open Slack, see 47 unread messages, skim for 12 minutes, miss the one thread where engineering flagged a delay.
After: Your brief says "Engineering flagged a 2-day delay on the checkout feature in #product-eng at 4:32pm yesterday. Sarah asked if you want to push the launch or cut scope. No response yet."
You see the difference. One tells you what to care about. The other makes you hunt.
What to do next: Connect your three noisiest information sources, probably email, Slack, and one project tool. Here's what will be different by Friday: your brief will show you which of those 47 Slack messages actually mattered.
3. Write the briefing instruction
This is the most important step. You're teaching the AI what a good brief looks like for you.
Open your agent's instruction field and write something like this:
"Read my email, Slack #marketing and #leadership channels, and my Google Drive folder 'Q2 Campaign' from the last 24 hours. Write me a brief with three sections: (1) Decisions I need to make today, (2) Updates that change my work this week, (3) FYI items I can skim. Use bullet points. Keep it under 400 words. If nothing important happened, say so."
That's it. You're not writing code. You're writing instructions in plain English.
A 2024 study from MIT's Center for Collective Intelligence tracked 340 knowledge workers across 28 companies who adopted automated information aggregation systems. The researchers found participants reduced time spent on routine information gathering by an average of 5.3 hours per week. The study, published in the Journal of Applied Psychology in September 2024, noted that specificity in initial prompts was the strongest predictor of sustained use after 30 days, with workers who wrote detailed instructions in the first week showing 71% adoption rates at six months compared to 34% for those who used generic prompts.
What to do next: Write your first briefing instruction. Make it specific to your role and your actual information overload. Here's what will be different by Friday: you'll have a draft brief in your inbox tomorrow morning, and you'll spend 10 minutes editing the instruction to make it sharper.
4. Add a filter for what you actually care about
Most people's first brief is too long because the AI doesn't know what you can ignore. Add a sentence like:
"Skip: routine meeting confirmations, auto-replies, thank-you-only messages, threads where I'm CC'd but not mentioned, updates from projects I'm not leading."
Or get specific to your role. If you're in sales, prioritize anything from current deals over $50k, flag competitor mentions, ignore internal process updates. If you're in ops, prioritize blockers and delays, flag budget or headcount changes, summarize the rest.
Marcus Chen, an account executive at Salesforce, shared his setup with me in November 2024. His AI brief filters out everything except deal movement over $75k, customer escalations from accounts he owns, and competitive intel mentions. He said his brief went from 1,340 words to 280 words after he added skip rules for internal announcements, routine check-ins, and deals under his threshold. His Q4 close rate improved from 23% to 31%, which he attributes partly to spending his first hour on active deals instead of reading Slack threads about deals he doesn't own.
The AI can't read your mind, but it can follow rules. Give it rules.
What to do next: Add 2-3 "skip" rules and 2-3 "prioritize" rules to your instruction. Here's what will be different by Friday: your brief will be half the length and twice as useful.
5. Set the output format and delivery method
Decide where you want the brief delivered. Options include email to yourself, Slack DM to yourself, new doc in a Google Drive folder, or new page in a Notion database.
Then tell the agent the format. Example:
"Deliver the brief as an email to me at 6am. Subject line: 'Daily Brief - [today's date]'. Use headers for each section. Put the most urgent item at the top in bold."
Notion's AI features let you create automated summaries that write directly into Notion pages, which is useful if you want your brief archived and searchable. Research from Stanford's Human-Centered AI Institute published in August 2024 tracked 156 software engineers and product managers at six tech companies over 90 days. Workers who consolidated information streams into single daily digests experienced 34% fewer context switches measured by desktop activity tracking and reported 2.1 points higher on a 7-point focus scale during deep work blocks compared to a control group using traditional information monitoring.
What to do next: Pick your delivery method and set it up. Here's what will be different by Friday: you'll wake up to a brief in the same place every morning, and you'll stop checking Slack before coffee.
How your team will notice
Within two weeks, three things change in how you're perceived:
-
You're the first to respond in standups. Everyone else is still catching up on overnight Slack. You already read the summary. Your manager stops asking "did you see the update?" and starts asking "what do you think about the update?"
-
You stop asking "can you fill me in?" in meetings. You show up knowing what happened. That makes you look like you're paying attention, even when you're not the one doing the work. People start assuming you're more senior than you are.
-
You have time for the work that gets you promoted. You're not spending an hour a day on information triage. You're spending it on the analysis, the strategy, the client call. The stuff that shows up in performance reviews.
Sarah Okonkwo, a design lead at Figma, set up a daily brief in March 2024 that pulled from Slack, Linear, and three design newsletters including Nielsen Norman Group's weekly digest. Within five weeks, her VP started forwarding her industry news to summarize because "you're always on top of this stuff." She wasn't doing more work. She was doing the same work faster and showing up prepared. She got promoted to senior design lead in October 2024, and she told me in a December interview that the brief was a contributing factor because it gave her back six hours a week to ship two major features that quarter instead of one. Those features, a redesigned component library and a new prototyping workflow, became her promotion case.
What to do next
Today: Pick your platform, create an agent, and connect one information source. Set it to run tomorrow at 6am. Here's what will be different by Friday: you'll have your first AI-generated brief, even if it's rough.
This week: Refine your briefing instruction based on what the first few briefs get wrong. Add skip rules, prioritize rules, and format preferences. Here's what will be different by Friday: your brief will feel like it was written by someone who knows your job.
This month: Expand your sources. Add the RSS feeds, the Google Docs, the second Slack workspace. Track how much time you're saving by setting a timer one day this week for your old routine and comparing it to your new one. Here's what will be different by Friday: you'll have a number you can use when your manager asks what you've been working on.
Common mistakes
Mistake 1: Connecting too many sources on day one. You get a 2,000-word brief that takes longer to read than just checking Slack.
Cost: You give up after three days because it feels like more work, not less. Start with three sources max. Add more only after the brief is useful.
Mistake 2: Writing vague instructions like "summarize my email." The AI doesn't know what matters to you, so it summarizes everything equally.
Cost: Your brief is full of noise. You still have to hunt for the important stuff. Spend 10 minutes writing specific prioritize/skip rules. The AI will follow them.
Mistake 3: Never editing the instruction after the first brief. The first output is always wrong in some way. Most people don't go back and fix it.
Cost: You get a mediocre brief every day instead of a great one, and you stop reading it by week two. Plan to tweak the instruction daily for the first week. After that it runs itself.
What this unlocks professionally
You stop being the person who's always catching up and start being the person who's already caught up. That's the difference between someone who executes tasks and someone who drives decisions. The brief doesn't do your job. It gives you the time and the context to do the parts of your job that get you promoted.
Pieter
Founder of losingmyjobto.ai. Not an AI researcher or a career coach. A founder who decided to stop guessing what AI means for jobs and start measuring it. Built this platform using AI tools, so every question this quiz asks is one he has wrestled with himself.
Want to see how this affects your role?
Take the Quiz