I've watched a lot of startups build video into their product the exact same way: treat it like a checkbox, ship it, and move on. That's the mistake.
Video isn't just something to watch; it’s a boatload of context and data.
Mux is the video infrastructure company that actually thought this through. Upload a video to Mux and you don't just get playback — you get building blocks your product can actually use:
Transcripts: Power search, summaries, and content discovery
Storyboards: Ready-to-use visual map of every video
Clips: Instantly extract any segment of a video
Thumbnails: Automatic, zero engineering time
Mux also stewards Video.js, the web's most popular open source video player. That's the kind of long-term ecosystem bet that most infrastructure companies aren't willing to make. (btw the Video.js v10 beta just dropped at videojs.org)
Patreon, Cursor, and Synthesia are all built with Mux. Get started free, no credit card required. Use code BIGDESKENERGY for an extra $50 credit.

Today we’re launching an MCP, making beehiiv the first newsletter platform you can run directly through ChatGPT, Claude, Gemini, or Perplexity.
For those of you who aren't as deep in the AI weeds or chronically online on X, let me explain what an MCP is and why it’s such a big deal.
MCP (short for Model Context Protocol) is an open standard that lets AI models like Claude connect to external tools and data sources (like Slack, Google Drive, and now beehiiv). You can think of it like an API, but supercharged: where APIs simply move data from point A to point B, MCPs layer in AI reasoning.
For example, using the Gmail MCP, you can simply ask your AI model of choice to summarize all of your unread emails, flag any that seem extra important, and craft draft responses for those that need to be replied to. All through the simple chat interface ChatGPT popularized a few years ago.
Until now, if you wanted AI to analyze your newsletter, it'd piece together an answer from whatever data it could find. It was only as smart as the snapshot you gave it.
The beehiiv MCP changes that entirely.
Instead of pasting data into a chat window, your AI connects live to your beehiiv account and can access, analyze, and act on everything inside it. For example, with the beehiiv MCP you can simply ask Claude to analyze your subscriber list and build a breakdown of the types and sizes of companies reading your newsletter.
Here’s what it uncovered about my most recent subscribers to this newsletter:

No dashboard or simple integration could do that. It requires context, reasoning, and the ability to connect the dots across large datasets.
So, why now?
As tools like Claude, ChatGPT, and Gemini become where people get work done, AI interfaces are becoming the new dashboard. Let me give you another example…
Linear is the project management tool that we use to track progress across all of our different projects and initiatives (think of it as an alternative to Trello or Asana). Historically, it’s been most heavily used by the product and engineering teams because it integrates directly with GitHub.
We chose to use Linear five years ago because it was beautifully designed, the UI/UX was super intuitive, and it was just a far superior product than the alternatives.
But they recently have leaned super heavy into agentic workflows and built an incredibly powerful MCP. Now our team rarely uses the web app: instead we create new tickets directly in Slack, and use Claude to perform analysis on all of the progress across our projects.

Leveraging Linear via Claude is 100x more powerful than the web app because our Claude instance also has access to:
GitHub: to understand the underlying code
Our database: to understand usage data and trends
Slack: to gather context from conversations and channels
Fellow: to extract transcripts and recordings from our meetings
In theory, Claude could automatically reprioritize features based on demand signals from sales call transcripts (and it can cross-reference the potential deal size via the HubSpot MCP).
What I find most admirable about Linear, however, is that they were already winning on product and UI/UX. But rather than falling into the innovator’s dilemma and avoiding the inevitable shift towards AI workflows, they leaned in and built something that totally contradicts their previous competitive advantages (the UI/UX of your app doesn’t matter when you’re doing work in Claude).
But Linear isn’t getting replaced by AI, it’s being fueled by it. They are just as important to our team than ever, if not more important. And while previously, it was mostly engineers and PMs using Linear, now the entire team from support to solutions leverages the Linear MCP.
Translation: they considerably expanded usage and seats.
So back to what we’re building at beehiiv: we are developing the underlying infrastructure for anyone with an audience.
Whether that's inside the beehiiv platform, through an AI, or whatever interface comes next — beehiiv is the operating system underneath all of it. That means leveraging beehiiv via Claude is 100x more powerful, for all the same reasons.

Today we’re releasing v1 of the beehiiv MCP, starting with read-only access. That means while chatting with your AI client of choice, you can ask things like:
Who are my most engaged free subscribers who haven't upgraded and what do they have in common?
Audit my website and give me a prioritized list of SEO fixes based on my content and structure.
Are there any unusual churn or unsubscribe spikes in the last 90 days? What posts or dates correlate with them?
But the real unlock is pairing beehiiv insights with actions in the other tools you already use:
Every Monday at 8am, pull last week's revenue across ads, subscriptions, and digital products and post a summary to Slack with a week-over-week comparison.
When a subscriber hits 10+ opens and 5+ clicks, create or update their contact in HubSpot and enroll them in an active sales sequence.
Every Friday, pull this week's send performance, summarize what worked, draft a Monday editorial brief in Gmail, and create a 9am calendar invite via Google Calendar.
Paying users can now request early access to the beehiiv MCP (request access here). We'll be onboarding users on a rolling basis… and everyone who gets in will also receive an invite to our private Slack community and dedicated MCP channel: home base for all early adopters.
We’re already deep in development on v2, which will introduce write-access and unlock capabilities like:
Pull my last 5 posts and generate a “greatest hits” roundup edition. Write it, format it, and save as a draft.
Build an automation for anyone who purchases a digital product: send a thank you email on day 1, a tips email on day 3, and an upsell on day 7.
Generate a new landing page for my paid subscription using testimonials from my best subscriber reviews.
This marks the latest chapter in our on-going effort to build the infrastructure to support anyone with an audience. Now, equipped with the power and capabilities made possible by the latest advancements in AI.
Two weeks ago at SXSW, I was on a panel with Courtney Johnson, a personal branding and social media creator. She said she spends 2 days a week creating content, and 5 days on admin.
I believe beehiiv has the opportunity to flip that on its head, for Courtney, and for creators and publishers around the world. We’re arming the rebels of the content economy.
“Claude, ask my readers to request early access to the beehiiv MCP. Make no mistakes.”
If you enjoyed this post or know someone who may find it useful, please share it with them and encourage them to subscribe: mail.bigdeskenergy.com/p/beehiiv-mcp


Credit: Me
I don’t ask for much, just jungle views and a desk cactus.
Think you can generate a better office? Reply with your submissions 📨.

Some of my favorite content I found on the internet this week…
beehiiv or Substack? I weigh in and chat with Dylan Byers from Puck on The Grill Room.
As a remote company that is built on Ruby on Rails, I’m a big fan of DHH. He joined My First Million and it was a great episode.
Ben Thompson dropped a banger about LLM paradigms, agents, and the AI value chain.
Maybe you’ve heard about the Delve scam controversy? If not, give it a deep dive here.

Chat with DenkBot — my AI clone. It’s trained on everything I’ve ever published and the entire beehiiv knowledge base 🧠.
Or you can book an hour session with me (real me) directly → here.

Turn on, tune in, drop out. Click on any of the tracks below to get in a groove — each selected from the full Big Desk Energy playlist.

Share this newsletter with your friends, or use it as a pickup line.
👉 Your current referral count: {{ rp_num_referrals }} 👈
Or share your personal link with others: {{rp_refer_url_no_params}}
📥 Want to advertise in Big Desk Energy? Learn More












