This project helped understand how to effectively make use of AI Agent orchestration frameworks like LangChain, how to implement human-in-the-loop AI agent systems, and how to validate and evaluate AI Agent outputs.
Project case study
CRM AI Assistant
Call-to-CRM AI Assistant reads a sales call transcript, suggests CRM updates, drafts follow-up tasks and an email, and waits for human approval before applying any changes. It is designed to make post-call CRM work faster without allowing the model to write directly to the CRM.
Key outcomes
- Accepts a pasted transcript, meeting notes, or a plain text upload.
- Loads the current opportunity from a seeded demo CRM.
- Extracts structured meeting facts and compares extracted facts against the current CRM record.
- Proposes evidence-backed CRM field updates. Drafts follow-up tasks and a follow-up email
- Pauses for human review and approval. Applies only approved changes