Create sophisticated conversational experiences using Amazon Lex

on YouTube ↗30m

" summary

Dropbox builds Dropbot with Amazon Lex, integrates it into the chat frontend and backend services, and launches it in October 2019 as an FAQ that evolves into a resolver of common issues. The bot initially solves 30% of issue types and grows to over 85% by expanding flows, using analytics, and storing transcripts. Lex features used include context in console, confidence scores and top‑5 intents, multi‑language support (10 languages), Amazon Connect integration, barge‑in and tunable silence/voice timeouts.

AI-written summary · about

" chapters (9)

1. 0:00Introduction and agenda
2. 1:29Why conversational interfaces
3. 3:43Lex journey and capabilities
4. 8:25Builder productivity themes
5. 10:11Stakeholders in bot building
6. 14:33Dropbox Drop Bot case study
7. 21:48Drop Bot demo and persona
8. 25:04New Lex features recap
9. 29:10Resources and conclusion

" related talks

• 47mBuilding sophisticated conversational AI experiencesChris Brown, Sandeep Srinivasan · AWS Summit San Francisco 2022
• 54melevating customer experience w/ conversational AIArjun Chatterjee · AWS re:Invent 2019
• 35mDesigning Experiences for ChatbotsIvan Pauletti · THE CONF 2017
• 54mTalk over text: Conversational interface design and usabilityPreston So · Frontend United 2018
• 31mSeamless Customer Service with Amazon Connect & AlexaAsif Mujawar · Conf42 Machine Learning 2021
• 32mIncrease productivity and satisfaction with an intelligent contact centerDave Rennyson, Shanthan Kesharaju · AWS Machine Learning Summit 2021
• 37mBots PanelPete Skomoroch, David Hall, Dennis Yang, Andrew Burgert, Sonal Gupta, Yishay Carmiel · AI By the Bay 2017
• 28mGive users relevant answers quickly with intelligent searchSara van de Moosdijk · AWS Machine Learning Summit 2021