In partnership with
Good morning 👋 This week’s stories all share a common thread: what happens when AI goes from assistant to operator. Whether it’s agents replacing support teams, coding platforms enabling solo founders, or automations cleaning your newsletter list, the line between human and machine execution is getting blurrier. Let’s get into it.
.
In this issue:
Salesforce cuts 4,000 as AI agents take over support 🥶
xAI launches Grok Code Fast 1, a cheaper and faster dev model 💨
Cursor shows how AI agents can ship production code 🚀
Replit’s CEO on using AI to build billion-dollar businesses 💰
The email automation that doubled our open rates ⬆️
How 433 Investors Unlocked 400X Return Potential
Institutional investors back startups to unlock outsized returns. Regular investors have to wait. But not anymore. Thanks to regulatory updates, some companies are doing things differently.
Take Revolut. In 2016, 433 regular people invested an average of $2,730. Today? They got a 400X buyout offer from the company, as Revolut’s valuation increased 89,900% in the same timeframe.
Founded by a former Zillow exec, Pacaso’s co-ownership tech reshapes the $1.3T vacation home market. They’ve earned $110M+ in gross profit to date, including 41% YoY growth in 2024 alone. They even reserved the Nasdaq ticker PCSO.
The same institutional investors behind Uber, Venmo, and eBay backed Pacaso. And you can join them. But not for long. Pacaso’s investment opportunity ends September 18.
Paid advertisement for Pacaso’s Regulation A offering. Read the offering circular at invest.pacaso.com. Reserving a ticker symbol is not a guarantee that the company will go public. Listing on the NASDAQ is subject to approvals.
News
Salesforce cuts 4,000 jobs as AI agents take over support
credit: Fortune
TL;DR: Salesforce just laid off 4,000 employees (nearly half of its customer support staff) as AI agents now handle a growing share of customer service. CEO Marc Benioff, who recently claimed AI wouldn’t lead to mass layoffs, is now aggressively pushing “agentic AI” to automate large chunks of Salesforce’s operations.
AI tools now handle customer support and are being deployed to tackle a backlog of 100M+ uncalled sales leads built up over 26 years.
The layoffs represent about 5% of Salesforce’s workforce, shrinking the support team from 9,000 to 5,000 employees.
Just two months ago, Benioff argued AI needed a “human in the loop” due to its ~90% accuracy—but has since pivoted.
While support roles are being cut, Salesforce is hiring sales staff to drive AI adoption, signaling a shift in job type rather than total headcount.
This is what the next five years of AI adoption might look like across enterprise SaaS. Leaders who once downplayed AI’s disruptive potential are now executing aggressive automation plans. Founders building in AI or selling into the enterprise should take note: AI isn’t augmenting teams, it’s replacing them in critical, customer-facing functions.
AI
credit: xAI
TL;DR: xAI just unveiled grok-code-fast-1, a purpose-built coding model optimized for speed, low latency, and tight integration with agentic development workflows. It’s fast, affordable, and already embedded in tools like GitHub Copilot and Cursor.
Trained on real-world pull requests and refined with partner feedback, grok-code-fast-1 excels at fast reasoning, terminal tool use, and context switching inside IDEs.
Scored 70.8% on SWE-Bench-Verified, but xAI emphasizes human evals over benchmarks—prioritizing real-world coding performance.
Supports TypeScript, Python, Java, Rust, C++, and Go; aimed at automating everyday developer tasks with minimal friction.
Pricing: $0.20/M input tokens, $1.50/M output, and just $0.02/M for cached input, making it one of the most economical daily drivers for agentic coding.
LLM coding agents aren’t just about raw power anymore—they’re about speed, cost, and frictionless tooling. xAI is quietly positioning Grok as the dev-friendly counter to OpenAI’s Codex and Anthropic’s Claude. If you're building dev tools or autonomous agents, this is the model architecture worth watching.
Productivity
Why Cursor’s AI agents are more than just autocomplete
TL;DR: In a live demo, Cursor’s VP shows how their AI agents automate core development tasks (bug fixes, code reviews, documentation) while integrating into workflows like GitHub, Slack, and CI pipelines. He notes that you must treat each agent as a task-specific teammate and manage context with surgical precision.
Cursor agents aren’t just code autocomplete; they independently analyze, write, review, and improve code with minimal human oversight.
Managing context properly (starting new threads for new tasks) is critical to maintaining output quality as complexity scales.
Developers can build custom AI commands to enforce style guides, flag anti-patterns, or auto-review PRs before merge.
Agents can be triggered from Slack or GitHub, auto-generating PRs and accelerating bug resolution, enabling “zero bug” policies.
Cursor is turning the IDE into an AI-powered command center, and it’s not just for coders. The rise of “personal software” means anyone can spin up AI-built apps for niche problems with near-zero marginal cost. But the demo makes something else clear: code is easy now; distribution is the new moat.
How 1,500+ Marketers Are Using AI to Move Faster in 2025
Is your team using AI like the leaders—or still stuck experimenting?
Masters in Marketing’s AI Trends Report breaks down how top marketers are using tools like ChatGPT, Claude, and Breeze to scale content, personalize outreach, and drive real results.
Inside the report, you’ll discover:
What AI use cases are delivering the strongest ROI today
How high-performing teams are integrating AI into workflows
The biggest blockers slowing others down—and how to avoid them
A 2025 action plan to upgrade your own AI strategy
Download the report. Free when you subscribe to the Masters in Marketing newsletter.
Learn what’s working now, and what’s next.
Founder Story
Replit’s CEO on building billion-dollar startups with AI (and why humans still matter)
TL;DR: In a wide-ranging interview, Replit founder Amjad Masad breaks down how AI is killing the “I can’t code” excuse. With platforms like Replit, non-technical builders can now ship real products fast. But while AI handles more of the grunt work, Amjad argues that success still hinges on domain expertise, creativity, and relentless execution.
Replit now supports 350,000+ active apps and is valued at $3B, showing the explosive potential of AI-assisted entrepreneurship.
AI lowers the barrier to entry but doesn’t replace the need for clear thinking, testing, and iteration.
Prompt engineering is the new literacy; precision matters when communicating with LLMs.
Engineers aren’t obsolete—just shifting toward roles in complex, high-stakes systems where AI still falls short.
Tools like Replit mark a shift from "learn to code" to "learn to build with AI." But Amjad’s take is a useful counterweight to the hype: this isn’t autopilot. It’s a new era where execution speed is amplified, but vision, domain knowledge, and grit are still what separate builders from tourists.
Tutorial/Framework
TL;DR: The MakerThrive team walks through how cleaning unengaged subscribers using beehiiv automation can dramatically improve newsletter performance. By cutting out cold leads (people who haven’t opened in 150+ days), open rates more than doubled.
Continuing to email inactive subscribers hurts your sender reputation and inbox placement across the board.
Use static segments to identify low-engagement subscribers and trace which acquisition channels are underperforming.
Set up automations to either re-engage or automatically remove cold subscribers—no manual cleanup needed.
Start with a one-email flow for fast impact; layer in multi-step sequences later for ongoing maintenance.
Dead subscribers don’t just skew your numbers—they make it harder for your real audience to see your content. Automating re-engagement and list hygiene isn’t optional if you care about growth—it’s baseline infrastructure for maintaining reach and relevance.
Want to sponsor this newsletter? This newsletter goes out to thousands of creators, builders, and early-stage founders every week. If you’d like to get your product or brand in front of them, you can book a sponsorship slot here
