Product Roadmap

Your assistant doesn't go dark when your laptop closes. Move seamlessly between local and cloud — same memory, same identity, same conversations.

Talk to your assistant out loud, interrupt it, and have it talk back — with low enough latency that it feels like a real conversation.

A native Apple-ecosystem channel for your assistant — texts, read receipts, the works.

Your assistant delegates to specialized helpers in parallel — research, coding, writing — and brings the results back to you.

Wire any service into your assistant with a simple webhook. If it can POST, your assistant can react to it.

I would really really like mobile/web support for locally hosted. through the vellum account. please!

I want my assistant to join my calls to take notes and have the context of the things Im working on. Ideally it can start tasks discussed in meetings before I tell my assistant to get on them lol.

Your assistant adapts to how you work — tone, habits, preferences, the people you talk to. It gets more yours the longer you use it.

backup of conversations is nice, and necessary for sure. But I accidentally started a scheduled 5 minute process that left about 27,000 "watcher-tick" conversations in the database, I don't need these anymore. I need a bulk way of cleaning these up without destroying the vector…

Allow the doctor to have longer more in-depth conversations, especially when it comes to SQLLite. The doctor in our situation kept "timing out" and resetting. It also ran out of conversation at a prime point for fixing, while investigating crashloop on sqllite - simply stating:…

Have my assistant write and read discord messages and act as a member of a community.

Join meetings, take notes, track action items, and pull them into your to-dos — without another tab to manage.

Recurring tasks and check-ins that don't break when an API token expires. Reliable background work, handled behind the scenes.

A work assistant and a personal assistant. Or one per project. Separate memories, separate identities, same you.

Your money, summarized and acted on. Categorization, alerts, and small automations you'd actually run.

Pull signals from the wearables you already have and turn them into useful nudges, not a dashboard you have to check.

Access your assistant from any browser, on any device, without an install. Same assistant, zero friction.

One assistant that coordinates multiple coding agents across your repos — planning, delegating, and shipping end-to-end.

Your assistant can talk to other assistants — coordinating schedules, sharing context, negotiating on your behalf.

A permissions model that's both safer and less annoying. Clear defaults, granular overrides, and no more guessing what you approved a week ago.

Your assistant notices when it got something wrong and quietly gets better at it. No fine-tuning pipeline required.

Add a skill that lets it construct code according to the best practices from Code Complete by Microsoft Press. One of the best "clean coding instruction books" I've ever read, including inheritance, functions, parameters, naming conventions, documentation, class files, and more.…

Would love to see new skills in the assistant, that teach it how to write proper scheduler jobs, and set the proper scheduler configurations, so that: it knows the difference between EXECUTE mode and the other modes (ie: script). Also, so it understands how to write, harden, chec…

Would love to see the Vellum assistant available as a self-hosted option on Linux. Currently the assistant is macOS-only for local hosting, which locks out a large portion of developers and power users who run Linux as their primary OS or on home servers. Use cases include: runni…

incognito mode where it won't make memories from the conversation and you can choose if memories to in or not