The previous chat experience struggled with adoption. Rigid, pre-mapped bot flows were exhausting to maintain and couldn't adapt to customer needs.
Connecting existing knowledge and AI workflows got teams up and running in minutes.
One of the pain points of the previous experience was cumbersome setup for deterministic bot flows. Guided logic for specific scenarios could be layered in as needed, and the whole experience was built on top of Front's AI pipeline so it scaled as the underlying models improved.
Creating a flexible information collection building block so the experience could adapt to be more guided when needed.
The information being collected could be defined in natural language and could cover any topic or scenario, so the pattern had to be scalable and adaptable.
As AI took on more, the requests that reached humans couldn't always be handled in real time.
Instead of leaving customers waiting by the chat window, we set clear response time expectations upfront, letting them know they'd hear back over SMS or email. Reply times adapted to each company's setup, with the flexibility to lean more toward live chat or async follow up.
Partnering with engineering to create a mock environment where both design and engineering could prototype, test and push updates.
I built a system that generated a palette from their brand colour and applied it consistently across every widget element, ensuring it always met accessibility standards.
Internal tooling