Sesame launches iOS preview, bringing its conversational agents out of the lab

Built by Oculus-era founders, Sesame is shipping a public iOS preview and framing its assistant as a collection of personal agents, with intelligent eyewear slated for 2027.

By · Published

Why it matters

A team that helped ship a new computing interface once before is now targeting voice-first personal agents, shipping software today while building hardware for 2027. If Sesame’s full-stack approach works, it could push assistants beyond chat windows and into everyday moments, raising the bar for how natural AI feels on the phone and, eventually, on your face.

Sesame launches iOS preview, bringing its conversational agents out of the lab

Sesame launched a public iOS preview of its conversational AI agents, an experience designed to feel less like a chatbot and more like talking to a person, according to TechCrunch (https://techcrunch.com/2026/05/28/sesame-the-conversational-ai-startup-from-oculus-founders-launches-its-ios-app/).

Sesame positions itself as a team of Oculus veterans now applying that hardware-software mindset to personal intelligence. Their previous work helped build Oculus, the VR company that sold to Facebook (now Meta), a deal that closed in 2014 (https://techcrunch.com/2014/07/21/facebooks-acquisition-of-oculus-closes-now-official/). Sesame’s thesis borrows from that lineage: make voice the most natural interface and build the stack end to end so the tech disappears into daily life.

What launched

The new iOS experience introduces Sesame’s agents to a broader audience. Rather than a single, general-purpose assistant, Sesame (https://www.sesame.com/) describes a "collection of personal agents" tuned for different moments and tasks, with an emphasis on fluid back-and-forth conversation. Sesame invites users to "think out loud, follow a thread, and discover something unexpected," reflecting a deliberate pivot away from prompt-and-response chat.

Alongside the mobile preview, Sesame is also running a research preview on the web (https://app.sesame.com/), giving early adopters and builders a window into how the system behaves as the team iterates. The mobile preview page (https://www.sesame.com/mobile-preview) outlines the company’s goals and how to get access.

The bet behind the app

Sesame frames its work as "personal intelligence with a point of view" that helps people do more of what they love. The team’s product philosophy, described on the site, centers on the nuance of voice as an interface and on tight integration of hardware, software, and machine learning. That full-stack posture shows up in the roadmap: Sesame is developing intelligent eyewear with high-quality audio for hands-free agent access, slated as "Coming 2027" on its homepage.

Shipping the iOS app now gives Sesame a daily surface where it can harden the agent experience while it readies hardware. If the eyewear lands as pitched, the app becomes both an on-ramp and a fallback, helping the agent feel ambient even without new devices.

How Sesame says it is different

From the materials the company has shared, Sesame’s focus is on conversation flow and personality, not just answers. The team describes research spanning conversational speech generation, personality modeling, and multimodality, supported by large GPU clusters for training and evaluation. That stack is aimed at making the agent feel present and consistent across contexts, whether you are tapping the app in a spare minute or speaking hands-free later.

There are no disclosed model details, pricing, or usage numbers in the available materials, and Sesame has not published regional availability specifics for the iOS preview. Sesame lists offices in San Francisco, Bellevue, and New York and presents itself as an interdisciplinary group of artists, makers, and engineers, reflecting its ambitions to blend product design with frontier ML.

Why this founder group

Oculus alumni have seen how a new interface can expand as the tech matures. With Sesame, they are aiming to cross the uncanny valley of voice: make an assistant that feels helpful, patient, and conversational without sliding into uncanny performance. Getting there is as much systems design as model tuning. The iOS preview is the first public look at how this team plans to thread that needle, with hardware on the horizon to make the experience truly hands-free.

Reader comments

Conversation for this story loads after sign-in.