homeProjectsResumeGitHub

Conversational AI Architect

Google CCAI

Designing and developing software tools and virtual agents for some of the world’s largest and most complex AI driven conversational user experiences.

A mixture of software engineering and client management to bring cutting edge new features of CCAI to market with Google customers.

A lot of development on internal tools, but we have open sourced this SDK for DialogFlow.

Managed:

  • Launching an Experiments initiative to iteratively improve virtual agents with metrics, AB testing and data driven approach.
  • Worked with R&D teams throughout Google to bring new technology such as Meena/LaMDA NLG to customer engagements.
  • Consulting to other teams on complex issues with DialogFlow CX Agent development and NLU tuning
  • Designing new methods to handle disambiguation, intent scoping and other conversation architecture topics.

Developed:

  • Tools to analyze and improve conversational logic: NLP intent, entity and conversation flows.
  • Data science to enable us to understand conversation flows and iteratively improve existing agents
  • Open sourced SCRAPI Scripting API and SDK for DialogFlow to make DialogFlow API functionality more easily usable by other developers.
  • Using full range of Google NLP and cloud tools for machine learning.
  • Cloud functions/webhooks to extend DialogFlow CX functionality.

ccai-flow.svg