Back to portfolio
Capability

AI Event Copilot — concept demo

This is an original, from-scratch recreation of the AI copilots I build for live enterprise events — the same patterns described in my AI/RAG case study, running entirely in your browser on fictional data. No backend, no API keys, no network calls.

AI Q&A Generation

Generates audience questions from a knowledge base, each with a grounded answer and a cited source — the moderator copilot pattern.

RAG Chatbot

Keyword-matches your question against the knowledge base and answers in a 'Based on [source]…' grounded style, with a friendly fallback when it can't.

Agent Mode

Plans a query, 'reads' a local dataset, and returns a conversational answer with an inline stat — the plain-language metrics agent pattern.

I build AI copilots for live enterprise events: RAG-grounded Q&A generation, knowledge-base chatbots, and agents that query event data on demand. This is an original, from-scratch recreation of those patterns using fictional data — built to show the capability, not a production system.

Generated from the TechConf 2026 knowledge base

Press Generate Questions to see AI-drafted questions and grounded answers.

Original concept demo · fictional data · runs entirely client-side.

Want the architecture behind the real thing?Let's talk →
In production this runs on AWS Bedrock — Knowledge Bases for managed RAG, Bedrock Agents for live metric queries, grounded in event data via a Kinesis Firehose → S3 pipeline.
Read the case study