Productivity app Hero announces an SDK that will complete your AI prompts for you

Creating the perfect router for AI chatbots is often a challenge, so much so that startups are creating roles for on-the-fly engineers. Consumer-facing AI applications are increasingly adding features such as automatically generated suggestion or recommendation buttons to entice customers to use the chatbot frequently and show them what the app can do.

Hero, a productivity startup founded by former Meta employees, announced a new autocomplete SDK today that will fill in prompts for you based on context. The SDK is currently available by invitation only and will allow developers to integrate this technology into their applications.

Here’s how it works: When you book a flight, you type “book a flight” in the prompt. The autocomplete SDK will start filling in fields such as To, From, On Date, At Time, Airline, and Return. You can stop at any point and send the query to the chatbot.

This is also useful when using an AI-powered photo or video creator, as autocomplete can help you with different parameters like object, style, location, landscape, and camera angle. Notably, Adobe’s new feature for creating audio clips with the Firefly app allows users to enter keywords in different sections of the prompt, such as mood, tone, and purpose, to easily create a ringtone.

Champion engineer Saharsh Vaidhi, who worked on the feature, said that you usually need a lot of work with the AI ​​application to get the results you want. With this autocomplete feature, you can get there with fewer prompts — or even just one.

Brad Kwok, co-founder of the startup, told TechCrunch that the company uses a series of models to predict what you might write in the next prompt.

“With AI autocomplete, we pull in all the inputs needed to complete an action, and get it done 10 times faster because there are fewer back-and-forths,” Kowalek said. “This opens up a whole new set of use cases ranging from travel to commerce, from advertising to customer support.”

Because there are fewer messages involved, companies operating at scale can save on server costs, he added.

Kowalek said he and his co-founder, Seung W. Lee, thought about building the technology based on their experience working on augmented reality features at Meta. In AR glasses, there are limitations on screen size, so the prompts interface should be simple – like adding parameters to a query.

The startup, which raised $4 million in seed funding last year, has secured $3 million in additional funding led by Forerunner Ventures. In the coming months, depending on the application and SDK growth, Hero plans to raise a larger round, Kowalek said.

Hero is testing a version of this technology in its app that helps users find time for a meeting or catch up with a friend through auto-complete prompts. The feature will be released to users within two months.

The company is also in talks with Koah Labs, an ad tech startup, to create AI-powered ads where brands can appear in autocomplete suggestions.