@bentossell of course, happy to!
When we started GoButler 18 months ago, our mission was to help people find products and services in the most intuitive way, using natural language. We thought that expressing your needs the way you think about it (“need a flight from la to nyc friday, coming back sunday. nonstop on jetblue) would make the process of finding products a lot easier. We started with a human-driven concierge, thinking that it would be the right way to go after the problem. Approximately six months later, we realized that while users really do see value in the simplicity of getting things done via text and in added layer of human intelligence, the inconsistencies this human layer added to their user experience kept them from using the service on a regular basis. The human layer added inconsistency in terms of both time (reply times for one and the same task at two different times fluctuated heavily) and in terms of quality (humans make mistake, they forget things, they give wrong information, etc.).
When we started working on this problem almost 18 months ago, there was little data/examples to be found on people expressing commercial needs in natural language - in all honesty, we couldn’t anticipate how much of the product would be automated (or not) in the long run, however, decided that the general need for simpler ways to get chores and any kind of transactions done was there. GoButler enabled us to gather valuable data that could help us build automated models in the future. We quickly saw how important response team/immediate feedback and product consistency were for consumers and realized that the real thing behind GoButler would be building smart AI that understands and translates natural language quires to structured and actionable intent.
Now, over the last year, we’ve seen millions of examples of people expressing their commercial needs in natural language. After all the “hard work”, thinking and building, we’re excited to be launching Angel.ai today. Angel.ai is our new product and the “new” focus of our company - we are developing cutting-edge, domain specific natural language processing models, based on the data that we’ve gathered through our concierge product, GoButler, and the learnings that we’ve been through. With Angel.ai, we want to make businesses’ inventories more accessible and customers’ product search experience a lot more efficient.
We're already working with a number of companies to enable their customers to interact with their inventory - whether on messaging apps, websites or within apps and are excited to see the first implementations go live soon. You'll see - being able to express things in natural language introduces a new level of speed and simplicity!
What, this is interesting. GoButler is now Angel.ai and is opening up its technology and "can process conversational requests across 40-plus commercial domains out of the box, providing businesses with an off-the-shelf solution to facilitate natural language-powered transactions."
“The objective with GoButler as a product was to figure out how people interact through conversational commerce, gathering millions of examples of how people express intent in a natural language message. From there, we used our human-assisted concierge to develop our AI and help train our models, aggregating the data from conversations to build out our data set,” Navid Hadzaad, CEO and co-founder of Angel.ai, tells TechCrunch.
I'll be diving into this (and related topics) with Navid tomorrow at the M1 Summit in New York, along with Thomas Smyth (Co-Founder & CEO of Trim), Thomas Hadfield (CEO of Prompt), Marah Lidey (Co-Founder & Co-CEO of Shine Text) mad Matthew Hartman (Director of Seed Investments at Betaworks)!
Love their mission but as an end customer it was disappointing. I tried their flight search feature and it didn't add any value. Doing the search on Kayak would have been more efficient.
@parisivy00 We don't have any consumer facing product implemented yet, so I'm wondering what you are comparing the Kayak experience to - if you are comparing it to the demo, it's really comparing Apples to Bananas as our demo is a pure demonstration of our NLP capabilities. I can guarantee, however, that there are many, many use cases where typing your intent out in natural language is going to be faster and simpler than translating that intent into a GUI.
Makerpad
Angel.ai
Kip
Father's Day Pop Up Shop
Raycast
Angel.ai