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.
Vercel shares results from integrating AI Agents into their customer support workflow, reporting a 31% reduction in support tickets. The company outlines three key pillars of their approach: an iterative, customer-centric methodology, thorough internal testing, and safe, incremental rollouts.
The post links to a deeper resource detailing how they achieved these results, positioning their experience as a practical case study for organizations looking to deploy AI in customer-facing support roles. The emphasis on gradual, tested deployment reflects a measured approach to AI adoption rather than a wholesale replacement of existing systems.
Vercel Developers announces an open-source AI Workflow Builder Template designed to let users visually construct durable, complex workflows and agents. The template supports AI-powered text-to-workflow generation, allowing users to describe desired workflows in natural language and have them automatically created.
The generated workflows compile into executable TypeScript using the Workflow DevKit (WDK), bridging the gap between visual design and production-ready code. This tool targets developers looking to build and orchestrate AI agents and multi-step workflows with a visual interface while maintaining the reliability of compiled, typed code.