Question Bot

The Question Bot series follows Andy Chen’s work to automate prospect Q&A across GTM. Each phase builds precision and trust: structuring Slack questions, grounding answers in code, and deploying cautious AI that delivers verified, human-quality responses without human escalation.

NOTE: Demo visuals use either blurred real data or synthetic placeholders to protect customer privacy.

Andy Chen

Builder

In this series

6 parts

Part 1

Question Bot - Grounded Answers

Sales teams often lose momentum chasing technical answers. Andy Chen, AI product manager, is building the AI Prospect Questions Bot to remove that drag. To that end, this solution helps sales teams craft human-quality answers without a human in the loop.

Part 2

Question Bot - Direct Confidence

Andy Chen’s latest demo of the AI Question Bot shows what happens when accuracy and restraint come together. Designed to remove subject-matter experts from every answer loop, the bot is now live in Slack, fielding real GTM questions and learning from expert reviews.

Part 3

Question Bot - Modular Precision

In this next phase of the AI Question Bot, Andy Chen moves from accuracy to adaptability. The system no longer operates as a single model. It now runs on modular components that can be tuned, tested, and trusted by experts across GTM and engineering.

Part 4

Question Bot - Decomposable Judgement

In the fourth demo of the Question Bot, Andy Chen, AI Native Product Manager, walks through the shift from one monolithic system to a set of small, testable components. The goal remains constant: answer general questions from prospects and customers without a human in the loop. This update focuses on accuracy, transparency, and trust in workflows where outdated or conflicting product information has historically slowed GTM teams.

Part 5

Question Bot - Direct to Customer

In Part 5 of the Question Bot series, Andy Chen, AI Native Product Manager at Abnormal, shifts the bot from an internal helper to a customer-facing bot. The goal is simple: shorten the distance between a customer question and a verified response, without adding load to Support or GTM.

Part 6

Question Bot - Closing the Customer Loop

In the sixth installment of Question Bot, Andres Lam, AI Native Product Manager at Abnormal, picked up where Andy Chen left off to harden the direct-to-customer experience on a customer-facing knowledge base. This demo focused less on retrieval mechanics and more on what it takes to run the bot day-to-day: UI, auth, fallbacks, and evaluation.

Keep exploring

Browse more workflows or follow other series.

All Workflows

Sécurité des identités et de l'IA

Plateforme

Ressources produits

Support

À la une

Leadership éclairé

À la une

Entreprise

Actualités et événements

À la une

Question Bot

In this series

Question Bot - Grounded Answers

Question Bot - Direct Confidence

Question Bot - Modular Precision

Question Bot - Decomposable Judgement

Question Bot - Direct to Customer

Question Bot - Closing the Customer Loop

Keep exploring