Builds made with OpenAI Dots

OpenAI's CEO says his dot is his favourite OpenAI product because it learns his workflow and style daily and takes on the chores that pile up as a gravity well of dread. A first-hand look at how a heavy user delegates.

Tibo Sottiaux set inbox zero as an active goal for his dot and, for the first time in his life, got there. A simple example of giving an always-on agent a standing objective instead of a one-off prompt.

Tibo Sottiaux asked his dot for a portrait, nudged it to search the web for a recent photo of him, and recorded himself browsing the dot's cloud computer to show how it worked.

Long-running VS Code virtual-pet project that added Codachi-style monster avatar designs for ChatGPT dots and Codex; shared on HN as 'ChatGPT Dot Avatars'.

Riley Brown called his ChatGPT dot for 45 minutes with his phone in his pocket and cleared the day's email and Slack tasks by voice. He used it as a hands-free chief of staff and is documenting the lessons for a video.

Antonio Leiva explains in Spanish how he used his always-on dot since the Friday before launch, connecting email and calendar and letting it propose work unprompted. A practical walkthrough of a real setup.

OpenAI researcher Noam Brown tested dots over a weekend and had his dot hunt down subscriptions to cancel. It then connected to a customer service channel and handled an entire texting conversation on his behalf.

Vox (@Voxyz_ai) switched to dots as his primary agent, praising the iMessage-style interface and desktop voice conversation that feels like talking to a person, and details the bugs he hit. A hands-on usage report.

Riley Brown spent a full day working through his dot, praising the voice experience and listing concrete gaps such as a better view of the Codex threads it spins up. A candid field report of what a daily dot setup feels like.

Paul Hudson woke to a cancelled train; his dot read the cancellation email, saw his ticket was valid on earlier services, checked live tracking and surfaced an alternative on its own. A concrete early-access story of proactive help.

OpenAI DevX engineer Dominik Kundel runs his personal dot as the front door to Codex, letting it handle follow-ups proactively instead of juggling threads himself. He argues dots and Codex complement each other rather than compete.

Derrick Choi, Codex APAC lead at OpenAI, describes his dot as basically an extension of himself that is available around the clock, posted with the launch-day demo video.

Kundel set up a brand-new dot and showed that it could immediately catch him up on things he had discussed earlier, drawing on his Codex and ChatGPT history while working on several tasks in parallel.

Simon Smith was stuck on a call when someone asked him to review a draft. His dot read the Slack message, found issues in the document and began adding comments at his direction.

Early-access community tester says his dot has been very helpful, liking a single main conversation while the dot coordinates other threads behind the scenes.

Early-access tester Dan McAteer watched his dot notice an unsent invoice, pull details from the email thread, draft it, and send it as a PDF after approval. The proactive behaviour was the standout of his testing.

A screenshot of Dominik Kundel's dot messaging him on its own to ask whether to follow a colleague's advice or record dot demos on Monday, an example of the proactive check-ins that define the product.

User shares his named dot 'Margaret', praising how proactive and fast it is, and requests the ability to embed things built with a dot or Codex inside ChatGPT Pages.

Browser game created entirely by a dot, shared as a sample of what the agent can build from a prompt.

A Pro user runs homelab projects through his dot and tries to have it delegate to Sol 6.1 Work chats, monitor progress and return when input is needed; he reports it started a task and finished setup work on a connected Windows VM.

A developer says dots removed the overhead of juggling many Codex threads while he builds enterprise intranet PHP, with the main annoyance being the auto-reviewer pausing his dot. Shows a real daily development setup.

A place where OpenAI Dot assistants can arrange playdates with each other, a lighthearted agent-to-agent experiment.

Laravel admin project that receives periodic automated updates committed by OpenAI Dots.

After plugging a dot into his work accounts, a user updated a Google Sheet with a different agent; the dot, watching email, flagged a possible duplicate in the badge data before production. A vivid example of proactive review.