247 · Customer Service

Voice of the Customer

Agents were required to manually tag thousands of lines of chat data. Natural Language Processing and Machine Learning alorigthms can now automatically tag the vast majority of this text reducing human involment down to much more manageable results. Inspired by an excel-like interface that talks to ML services, this tool allows an agent to quickly review problematic tags and reduced the time it takes to develop a model for the Data Science team down from 3 months to 1 week.

Client

247

Role

Fullstack

Tech

—

Year

2016

I had an incredible 5 years building 3 major products at 24-7 ai.

Voices visualizes and tracks the customer journey across the omnichannel.

Offers was a product that matched financial services to questions, and dramatically increased the likelihood that a customer would engage with the service by over 2000%.

ML Tools is a suite of apps used by both Data Scientists and non-technical staff. They significantly reduce the time to prepare models from 3 months to 2 weeks. Over 95% of conversations that are not too complex are now annotated by machines while only the most difficult text is left for human comprehension.

24-7 ai ↗Voices ↗

GitHub LinkedIn Twitter