Sesame launches iOS app with conversational agents designed to think out loud

Sesame, the conversational AI startup founded by former Oculus leaders, launched its iOS app on May 28 with four distinct agents designed to prioritize natural dialogue over response latency. The app is available in 39 countries at no cost and represents the company’s first major step toward agentic capabilities and future hardware integration planned for 2027.

Rethinking Conversational Flow

According to TechCrunch AI, Sesame addresses a fundamental tension in AI chatbot design: users expect rapid responses, yet slower, more deliberate replies tend to be more accurate. Rather than choosing one extreme, the startup built systems that allow its AI agents to search and retrieve information in parallel while speaking, enabling mid-sentence pivots when new context emerges. This mimics how humans naturally adjust their responses mid-conversation when remembering additional details—a departure from the turn-taking structure that defines most current LLM interfaces.

The four agents—Maya, Miles, Simone, and Charlie—each carry distinct voices, personalities, and persistent memory across conversations. During Sesame’s research preview phase, the first two agents attracted over one million users within weeks of the beta launch, signaling early market interest in the conversational-agent category.

Feature Set and User Feedback Integration

Sesame’s iOS app incorporates features refined through beta testing, including search cards with image results for visual concept exploration, a notes function for capturing key takeaways, and a texting mode for situations where voice interaction is impractical. An incognito mode allows agents to access prior conversation context without persisting data to memory—a privacy-conscious design choice. Deep-dive functionality enables users to request more granular, in-depth results without repeatedly re-prompting.

The company, which raised $250 million in Series B funding from Sequoia and others, views this iOS release as foundational technology. Per TechCrunch, Sesame has signaled that agents will eventually move beyond conversational interaction to execute actions on behalf of users, justifying the “agent” terminology over “chatbot.”

Why This Matters

Sesame’s emphasis on conversational naturalness over latency optimization addresses a real friction point in how enterprise and consumer users interact with AI—most LLM applications treat thinking as a hidden cost to be minimized rather than a feature of dialogue. If the parallel-search-while-speaking architecture proves durable and scales beyond iOS, it could influence how teams building voice-forward AI interfaces design user experience. The roadmap toward intelligent eyewear in 2027 also signals that the startup is betting on wearable form factors as the primary interface for long-form conversational AI, a thesis that hinges on whether hands-free, eyes-free interaction becomes the dominant paradigm. Early adoption metrics from the research preview suggest the concept resonates, but sustained engagement post-launch will determine whether Sesame’s approach is a lasting UX innovation or a feature that differentiates only within a narrow cohort of power users.