Say you have a job with a big company and you need to know just how much getaway time you have left, or how to add your new infant to your health care. This normally includes emailing or calling HR and awaiting a response, or it could even include crossing multiple systems to get what you need.
Leena AI, a member of the Y Combinator Summer 2018 class, wants to alter that by constructing HR bots to answer question for employees instantly.
The bots can be incorporated into Slack or Workplace by Facebook and they are constructed and trained using details in policy documents and by pulling data from different back-end systems like Oracle and SAP.
Adit Jain, co-founder at Leena AI states the company has its roots in another start-up called Chatteron, that the founders began after they left college in India in 2015. That item helped people build their own chatbots. Jain states along the way, they discovered while doing their market research study, an especially strong requirement in HR. They started Leena AI last year to resolve that particular requirement.
Jain states when constructing bots, the group found out through its experience with Chatteron, that it’s much better to focus on a single subject since the underlying artificial intelligence model gets better the more it’s utilized. “Once you produce a bot, for it to really to add worth and be [incredibly] precise, and for it to truly go deep, it takes a great deal of time and effort and that can just take place through verticalization,” Jain explained.
Picture: Leena AI What’s more, as the founders have ended up being more knowledgeable about the needs of HR, they have actually found out that 80 percent of the questions cover similar topics like trip, ill time and cost reporting. They have also seen companies using comparable back-end systems, so they can now build standard integrators for typical applications like SAP, Oracle and Netsuite.
Obviously, even though people may ask comparable concerns, the business might have distinct terminology or individuals might ask the question in an unusual method. Jain states that’s where the natural language processing (NLP) comes in. The system can learn these variations over time as they develop a larger database of possible questions.
The business just launched in 2017 and currently has a dozen paying clients. They want to double that number in simply 60 days. Jain believes being part of < a data-entity="y-combinator" data-type="company" href="https://www.crunchbase.com/organization/y-combinator/" target =" _ blank" > Y Combinator ought to help in that regard. The partners are assisting the team improve its pitch and making introductions to business that might make use of this tool.
Their ultimate objective is absolutely nothing less than to be ubiquitous, to help bridge multiple legacy systems to provide responses seamlessly for workers to all their questions. If they can attain that, they must be an effective business.