Pavel Samsonov

Self-service chatbot design tool

A major Canadian financial institution requested my team's help with using generative AI to optimize their internal operations. We designed a platform that provided a repeatable approach compliant with their governance requirements.

Product impacts

Helped double assets under management by increasing the productivity of customer-facing agents and their support teams, who were now empowered to solve problems visible only to front-line staff, as well as iterate on them rapidly.
Reduced dev time to minutes compared to months of effort in the past, as customers with an identified GenAI use case no longer had to chart their own course through the complexities of IT tickets and governance approvals.
40% decrease in I.T. requests by allowing users to define their own bots with existing components that already had governance approval; this capability would go on to be extended to all AWS customers as Amazon Q for Business.
75% of support queries automated by generating a self-help response from the existing - though disorganized - knowledge base, allowing support staff to focus on higher-value tasks.

Empowering financial services teams to innovate independently

The customer’s executive leadership set aggressive annual targets, and teams looked to Generative AI as a way to increase their productivity. IT and compliance orgs were inundated with requests for implementing AI apps. My team was asked to define the value case for a self-service platform, and design the intranet experience around it.

Identifying enduring needs
  • Business users
  • IT leadership
  • Governance teams
Building the Value Case
  • Executive alignment
  • Amazon PRFAQ writing
  • Journey mapping
  • Solution hypothesis

We found that internal thought leaders who had previously been successful at improving their team’s productivity through AI apps focused on one well defined use case and iterated frequently to tailor the app towards the needs of their unique context. Through a series of workshops, I aligned business and technical stakeholders around this strategy.

Iterating through artifacts
  • Service blueprint
  • Wireframes
  • Functional PoC
Validating with customers
  • Line-of-Business leaders
  • Super-users
  • Governance teams

We developed a fully functional proof of concept in a one-day Experience Based Accelerator and tested it directly with customers throughout the event to deploy 15 data retrieval and transformation apps in under 5 minutes. I defined the research strategy to gather data from this exercise, and led the synthesis activities afterwards to identify gaps in the workflow, design the future state, and set the roadmap towards General Availability.

Get in touch if you want to learn more about this case study.

upstream workflow
  • browse and search apps
  • clone existing apps
  • access governance
  • prompt engineering
downstream workflow
  • test and refine app
  • set access rights and publish

Get in touch if you want to learn more about this case study.