Back to projects

B2B mentoring and learning platform

FutureLab AI Workflows

AI-assisted transcription, summarization, and workflow automation for mentoring operations and B2B delivery.

Outcome

Helped product and operations teams turn session data into summaries, follow-ups, and delivery visibility.

Problem

Mentoring sessions produced valuable context, but too much of it disappeared into manual notes and follow-up work.

Approach

Built service objects around transcription and summarization jobs, added tests around edge cases, and treated AI output as reviewable product data rather than magic text.

Architecture

Rails workers process recordings and session metadata, call model services, persist generated summaries, and expose review states inside existing admin workflows.

Result

Introduced practical AI workflows into production while keeping the platform maintainable for a small engineering team.

Lessons learned

The best AI features often look like boring workflow software: queues, retries, review states, and clear ownership.

Constraints

  • Integrate AI into an existing Rails product without destabilizing core flows.
  • Keep outputs reviewable for operations teams.
  • Control cloud and model costs as usage grows.

Technical decisions

  • • Wrapped model calls in tested service boundaries.
  • • Used asynchronous jobs for long-running transcription and summarization.
  • • Optimized AWS resources and database access patterns alongside feature work.

Key features

  • • AI-generated session summaries.
  • • Operational review flow for generated content.
  • • Cost-conscious background processing.