SeedLoop AI: An Autonomous Multi-Agent Framework for Intent-Driven User Discovery and Day-0 Community-Led Growth

Project

PFE-07 — Growth, Multi-Agent AI, NLP, Growth Engineering

Build a multi-agent system that identifies a startup’s first users and turns community feedback into a product roadmap.

Objectives

• Discover people describing a relevant pain point and rank them by intent.
• Create a personalised quiz or prototype for each prospect.
• Orchestrate Slack and Discord communities, including moderation and introductions.
• Convert feedback into a prioritised backlog traceable to source conversations.

Key challenge

Turn messy and contradictory chat conversations into a backlog that a founder would build.

Target

Match a human product manager’s backlog using rank correlation and top-10 precision/recall.

Tech stack

Python, Next.js, LangGraph, Slack API, CrewAI, Discord API, LLM APIs, PostgreSQL, Qdrant.

Profile

Student in software or AI engineering.

Format and location

• 1 intern
• Full-time, on-site internship
• 4 to 6 months
• Knowledge Place, Monastir, Tunisia
• Market: Global startup founders

General requirements

• Willingness to learn and take on the challenge
• Relevant skills and interests
• Adequate written and spoken English
• Ability to work alongside people from diverse backgrounds and disciplines
• Availability for long-term work after the internship