Case Study · PitchFabrice

AI-Powered Startup Due-Diligence Interviewer

For a leading venture capital firm, the early stage of founder screening is among the most time-consuming parts of the investment process. Every week, dozens of founders send in pitches — each requiring a structured conversation, a set of consistent diligence questions, and a clear written summary for the investment team to act on.

AI Application DevelopmentVoice AIGenerative AIWeb App DevelopmentMulti-Agent Systems
Pitch Fabrice interface preview across devices
Overview

The project

We built Pitch Fabrice to handle that entire first stage automatically. It is a voice-first AI platform that conducts real-time conversational interviews with startup founders, asks the right questions based on their stage and business model, and generates a structured investor-ready debrief report after every session — ready for internal review without any manual note-taking or follow-up calls.

The challenge

What we had to solve

Consistent, Stage-Aware Interviews

Early-stage startups require different diligence questions than growth-stage companies. A single rigid script would not work. The system needed to adapt intelligently to each founder's context.

Real Conversations, Not Surveys

Founders needed to feel they were having a genuine conversation, not filling out a form. A stilted or mechanical experience would undermine trust and produce shallow, rehearsed answers.

Automated Report Generation

After each interview, the VC team needed a clean, structured debrief report covering company summary, fundraising details, traction, and stage-specific insights — without anyone having to write it manually.

Session Continuity

Interviews are often interrupted or rescheduled. The platform needed to support pausing and resuming sessions without losing any context or progress. —

The solution

What we built

We designed Pitch Fabrice as a multi-agent voice platform, with a supervisor AI at the centre orchestrating a pipeline of specialised agents — each built for a specific founder stage: Early Stage, Growth Stage, Confirmation, and Pitch Deck Analysis.

When a founder joins a session, the system identifies their stage and routes them to the appropriate agent, which conducts a structured conversational interview across founder background, business model, market size, traction, and fundraising terms. The conversation flows naturally through OpenAI's Realtime API, making it feel unhurried and genuine.

Before the interview begins, founders upload their pitch deck as a PDF. The system uses vision AI to extract and cross-reference key facts, so the agent already knows the basics before the first question is asked — and can probe deeper where it matters most.

After the session closes, the platform automatically generates a structured debrief report covering company summary, fundraising instrument classification (SAFE, convertible note, priced equity), traction data, and stage-specific insights. Sessions are fully persistent, backed by MongoDB, so founders can pause and resume at any point without any loss of context.

Inside the product

A voice-first interview, and the debrief it writes for you

pitchfabrice.fabricegrinda.com
End Pitch
Hey there! I believe you have a startup idea to pitch me. I’ll ask you 30 to 35 questions and give actionable feedback. To start, share your name, email, a one-line description, and upload your deck if any.
Hi, I’m Alex Chen. I’m building BrightLoop, an AI assistant that turns customer-support chats into product insights for SaaS teams.
That’s a clear problem to solve. Let’s start with you, tell me about your background and the genesis of this idea.
+Ask Anything🎤
This pitch is for informational purposes only and not a binding offer
Real-time conversational interview, voice or text
pitchfabrice.fabricegrinda.com/debrief

Investor Debrief · BrightLoop

Early stage

Company summary
AI assistant that converts customer-support conversations into structured product insight for SaaS teams.
Fundraising
SAFE
Instrument
$1.2M
Target raise
$8M
Post-money cap
Traction
12 design partners · $8K MRR · 30% MoM growth.
Stage-specific insights
Clear pain pointStrong founder-market fitValidate willingness to pay
Auto-generated after every session, ready for review

Voice interviews

Natural real-time conversation, not a form.

Multi-agent

A supervisor routes each founder to the right agent.

Deck analysis

Vision AI reads the pitch deck before it asks.

Auto-debrief

Structured investor report after every session.

Resume anytime

Pause and continue with full context intact.

Screens reconstructed from the live product. Debrief figures are illustrative.
Key features

Highlights

Conversational Voice Interviews

Founders engage in a natural voice conversation with an AI agent that asks structured diligence questions across founder background, business model, market size, traction, and fundraising terms.

Multi-Agent Orchestration

A supervisor AI coordinates a pipeline of specialised agents using the OpenAI Agents SDK, routing each founder to the right question set based on their startup stage.

Pitch Deck Analysis

Founders upload their deck as a PDF. The system extracts and cross-references content using vision AI, pre-filling known facts before the interview begins so questions are sharper and more relevant.

Automated Debrief Generation

Post-session reports are generated automatically, covering company summary, fundraising instrument classification, traction data, and stage-specific insights — ready for internal VC review without any manual effort.

Session Persistence and Resume

MongoDB-backed session state with full context replay, allowing interviews to be paused and resumed at any time without any loss of progress. —

Tech stack

Under the hood

Next.js (TypeScript)OpenAI Realtime APIOpenAI Agents SDKMongoDBAWS App RunnerDocker
Business impact

The outcome

Pitch Fabrice significantly reduces the manual effort of initial founder screening, enabling the VC team to process more deal flow with consistent, structured data captured from every conversation. What previously took hours of calls and note-taking now happens automatically, so the investment team can focus their time and energy on the decisions that actually require human judgment.

*End of Case Study — Pitch Fabrice*

Have a project like this?

Tell us what you’re building. We’ll come back with a clear, fixed-scope plan to ship it.

Book a free call

Wait! Don’t Take Off Yet... 🚀

Let us guide your next big move!
1. Custom Project Roadmap
2. Pricing Estimate
3. Completion Schedule
Simply fill out the form and we’ll get in touch with your FREE consultation!
Book a free call