Calendly's Playbook: Treat AI as Core Platform Infrastructure

Calendly's product strategy is spearheaded by Chief Product Officer Stephen Hsu, who joined in May 2023 after leading the $1.8 billion Integration Platform and API Management organization at Salesforce's MuleSoft. His background in API management, automation, and data integration directly informs Calendly's push to embed AI as a core platform function rather than a set of features. The company's AI-powered tools, such as the Notetaker for summarizing meetings and an AI Assistant for conversational queries, are built using large language models from providers like OpenAI. Calendly's policy is to send only the minimum data necessary for a feature to function and prohibits third-party providers from using customer data to train their models. This "observability-by-design" approach is critical in AI platform development, embedding monitoring at the earliest stages rather than retrofitting it post-deployment. For platform teams, this means defining AI-specific metrics for model drift, data quality, and even token usage costs, moving beyond traditional application performance monitoring. AI observability requires a triad of metrics tracking performance, cost, and quality to define system health. In the logistics sector, this platform-centric AI approach is mirrored in the shift from transactional APIs to AI-enhanced APIs that create self-optimizing ecosystems. These modern APIs automate complex workflows, like rerouting shipments during a disruption, by analyzing real-time data from traffic, weather, and warehouse capacity. This enables logistics providers to move from merely reporting a delay to proactively and autonomously resolving it. The global AI infrastructure market is projected to grow from approximately $26 billion in 2024 to over $221 billion by 2034, driven by massive investments from major tech firms and the soaring computational demands of generative AI. This rapid expansion is creating a competitive divide between companies treating AI as core infrastructure and those still in experimental phases. For engineering leaders, this infrastructure-first approach shifts team responsibilities. Data scientists might own model quality metrics, security teams manage safety and bias monitoring, and platform engineers become responsible for the underlying infrastructure performance and reliability. This structure necessitates a unified data architecture, as former Calendly CPO Steven Shu noted that clean, structured data is more critical than algorithm sophistication for AI success. Investor focus is expanding from AI software applications to the infrastructure powering them, including data centers, semiconductor manufacturers, and cloud providers. Spending on AI data centers alone is expected to exceed $1.4 trillion by 2027, with C-suite executives, not just IT departments, now driving these strategic investment decisions.