A Self-Driving Computer

At Amna, we’re experimenting with the fastest way to get things done on a computer.

A faster car doesn’t always get you to your destination sooner, sometimes you need a faster route. We’re going to find that route for you.


Self-driving cars are exciting! We set our eyes on the destination, and the car helps us get there.

To get there, they need two things- underlying mapping software that tell them where they’re going, and onboard computers to figure out items on the fly.

“Self-Driving” computers, like our PCs, are similar. We set a task, and they help us accomplish them. Whether it’s creating a business plan, planning a trip or scouting leads, our computer should be able to help you get there. And the number of clicks for you to accomplish a task should be minimal.

We’ve been dreaming about these systems since forever. Here’s a snippet of Project Aura from CMU Researchers in the early 2000’s.

Minus the anachronic sound effects, it’s eerily similar to the world we might envision with Siri or Google Assistant.

A computer actively queries, consults and gets information to accomplish a specified task on behalf of the user.

But did you notice that there was an interface around every corner? One on the coffee machine?Another on the wall? An interface suited to each context. The interface morphs itself to suit its unique location.

And that’s what computers lack today. Context. Your PC lacks the awareness to understand what it is that you’re working on. Here are your suggested apps. Here are your suggested files. But really, what you’re trying to accomplish is “complete filing your receipts”.

The GUIs of today don’t understand our larger intent. And even if they did, we don’t have a good way of telling them what we’re working on.

Our Approach

We are constructing a better map to teach computers how tasks are accomplished.

Are your 30 Chrome tabs completely random? Not quite. A few tabs are to find a plumber. Another few are for leaving comments on a document. And a couple more to find a birthday gift for a friend. And that’s why all those tabs are meaningful. Because they add up to something.

That’s also where systems of today fall short. Connections are made based on similar content. And based on that content, we’re going to guess your intent. And that’s shady. It’s why targetted advertising feels wrong. “I didn’t tell you what I was doing”

The Amna approach, reverses that principle. We make it easy to say what you are working on, your intent, and then let you pull in all of the supporting content. That’s our map. It’s more robust because there’s no guess work here. We can understand the way you accomplish your task to optimize it for the next time.

Using that map, our underlying engine optimizes a user’s behavior to achieve their task. This can include suggestions, introducing new services, or introducing completely new workflows.

What Changes?

  • Easy App Discovery

    • The system finds the right service for the right job. Are you researching for an essay, the system can find a suitable service to summarize articles for you. Are you finding flights? Great, here’s a new website that does it well. And all this happens, right in the context you want it.
  • Goodbye UI

    • Does it matter if you book your next trip through Expedia, Google Flights or Orbitz? If you take away the user experience - decisions usually come down to price and routing. A computer-driven interface strips away notions of UI and lets services focus solely on the data they provide.
  • Adopt a Process

    • Most of us have searched for templates, sample letters and examples of ways to do things. What if there was a template for everything related to job-searching? You could adopt an entire process, rather than just fragments. In fact, you could sell an entire process, rather than just a fragment.

Where we’re going - as long as we know what we want to do, the computer does the rest.

Privacy and trust are super important for us. That’s also why we are really transparent and open with our intentions! Love what we’re saying - get in touch.

See our culture and values.

comments powered by Disqus