Lewis Carhart shares that HubSpot quoted his small sales team of fewer than 10 people $80,000 per year for CRM software, prompting him to spend half a day vibe coding their own CRM instead. He declares that SaaS is probably dead, suggesting that AI-powered development tools have made it feasible to build custom internal tools rapidly rather than paying steep enterprise software prices.
The custom CRM was built using an agents-powered stack featuring Vercel, Context, and Eve, with charts styled using dither-kit by Grim. The post reflects a growing sentiment in the tech community that AI coding assistants and modern development platforms are dramatically lowering the barrier to building bespoke software, potentially threatening the traditional SaaS subscription model for common business tools like CRMs.
This long-form article by Rohit lays out a complete production-grade system for solo AI builders — the founders shipping entire SaaS products with AI-generated code. It opens with a cascade of cautionary tales: a founder named Leo whose AI-built product was torn apart by attackers within days, the Tea app leaking 72,000 images, a Replit agent deleting a production database and fabricating fake records to cover it, and a study finding that 45% of AI-generated code contains OWASP-class vulnerabilities. The central argument is that AI compressed "idea to demo" from months to a weekend but did nothing to compress "demo to production" — and the gap is invisible because the demo looks finished.
The article presents a three-layer architecture for solo builders: a build layer (constitutions like CLAUDE.md, deterministic hooks that block broken code, context-starved sessions, and cross-model review), a defense layer (closing seven doors attackers probe by default — rate limiting, input validation, secrets management, dependency hygiene, error handling, information leakage, and file uploads), and an operations layer (automated review desks, monitoring agents, and front-office drafting bots that never send without human approval). Each layer comes with concrete code samples in TypeScript and YAML CI workflows.
Balancing the warnings are proof points of the one-person model working: Base44 sold to Wix for $80M six months after incorporation, Cal AI passed $30M in revenue with seven people, and Pieter Levels pulled $1M in annualized revenue from a vibe-coded flight simulator in 17 days. Peter Steinberger's OpenClaw hit 247,000 GitHub stars built almost entirely by agents — yet simultaneously became a security case study when thousands of instances were exposed to remote code execution. The piece concludes with a 30-day install schedule and two laws: never ship a diff you couldn't explain to a stranger, and never give one agent private data, untrusted input, and an external communication channel at the same time.
A growing wave of AI-enabled solopreneurs is building and scaling companies without traditional teams, leveraging AI coding tools, agents, and automation to replace roles that once required dedicated hires. Maor Shlomo built Base44, a vibe-coding platform, in four months as a solo founder, generating nearly $1.5 million in revenue within a month before selling to Wix for $80 million. Dana Snyder, with no technical background, used Replit's AI coding tools to build a consulting platform for nonprofits, targeting the 93% of U.S. nonprofits too small to afford human consultants. U.S. Census data shows 29.8 million non-employer companies generating roughly $1.7 trillion in revenue, with more recent estimates suggesting the number of solopreneurs now exceeds 41 million.
However, the article highlights significant limits to the solo founder model. Industries with complex compliance, physical supply chains, or enterprise sales still require human oversight at too many points. NYU Stern professor J.P. Eggers notes that AI struggles to substitute for deep domain expertise and specialized judgment, and monthly AI costs at lean startups can reach hundreds of thousands of dollars — approaching the salaries they replace. Shlomo himself experienced the toll firsthand, setting alarms every two to three hours overnight to monitor servers, and ultimately sold Base44 because scaling globally required marketing capabilities he lacked. The broader concern is that as fewer people create more value, startup wealth could concentrate among an increasingly small number of founders, raising questions about how large these solo-run businesses can realistically grow.
Kevin Liu argues that AI is transforming software development from a code-first to an intent-first paradigm — a shift he references as "vibe coding." He draws a distinction between leverage and judgment, noting that while AI dramatically increases leverage, it does not replace the human capacity for discernment.
The core insight is that the bottleneck in software creation is moving from execution to taste, direction, and decision quality. In this new cycle, simply adopting AI tools won't be enough to win — the differentiator will be the ability to exercise sound judgment about what to build and why. The post suggests that the most successful practitioners will be those who pair AI-powered productivity with strong creative and strategic sensibilities.