Menu
FJ Labs
Industry : Artificial Intelligence | Digital Media | Personal
Tech Stack : AWS Serverless (Lambda, Step Functions) | OpenAI | Microsoft Graph API | MongoDB | React.js
FJ Labs runs one of the most active deal pipelines in venture capital. That means a constant flow of investment proposals, SAFEs, convertible notes, and equity agreements arriving every single day, each needing to be read, understood, extracted, and moved through an approval process before the team can act. When your deal volume is that high, the documents stop being a manageable administrative task and start being a bottleneck that slows everything else down.
We built a platform to remove that bottleneck entirely. The moment a document arrives in the team’s inbox, it is captured automatically, read by AI, and moved into a structured workflow without anyone needing to touch it. Every key term is extracted, every stakeholder is notified at the right moment, and every decision and comment is tracked in one central place. The team’s job became reviewing and deciding, not managing the paper.
The challenge was not just volume. It was building something accurate and trustworthy enough that a world-class investment team would actually rely on it.
SAFEs, convertible notes, equity agreements, and investment proposals arrive on a rolling basis. Manual review of each one was creating real delays in a process where speed genuinely matters.
Investment documents contain complex legal and financial language, variable structures, and critical terms buried in dense paragraphs. The AI needed to extract the right information accurately.
After extraction, each document had to flow through a defined multi-stakeholder approval process, with the right people notified at the right time and a clear record of every decision made.
With multiple reviewers, admins, and stakeholders involved, the team needed a single source of truth where they could see every document’s status, every comment, and every decision, in one place and in real time.
Our development team designed and deployed FabriceAI, a robust AI-powered web application tailored to present Fabrice’s viewpoints through a seamless, engaging user interface.
We built the platform around three things that had to work perfectly together: intelligent document capture, accurate AI extraction, and a self-running approval workflow. Documents come in through Microsoft Outlook, and a direct integration means nothing gets missed.
The moment a document is captured, the AI layer goes to work, reading through the content and pulling out structured investment data including key terms, instrument type, company details, and financial conditions, consistently and accurately, every time.
From there, the approval workflow takes over automatically. Stakeholders are notified by email at each stage, sign-offs are collected, and the document moves forward without anyone having to chase it.
A centralised dashboard keeps the whole picture visible: every document in the pipeline, its current status, who has reviewed it, and what decisions have been made. Role-based access means reviewers, admins, and super-admins each see exactly what they need to see.
Documents arrive from email and are captured automatically through a direct Outlook integration, with zero manual upload or filing required from the team.
The AI layer reads complex legal and financial documents and extracts structured investment data, including key terms, instrument classification, and financial conditions, accurately and consistently.
Once extracted, documents move through a defined multi-step approval process automatically, with email notifications and sign-off collected at every stage.
A real-time web portal gives the investment team a complete view of every document in the pipeline, including status, reviewer commentary, and decisions, all in one place.
Reviewers, admins, and super-admins each interact with the platform according to their role, keeping the workflow appropriately controlled without extra complexity.
Every action, decision, and annotation is captured permanently, giving the team a complete and reliable record for compliance, reference, and accountability.
The time from document receipt to investment decision shortened significantly. Manual data entry disappeared. And the team gained something they did not have before: a single, always-current view of every deal in the pipeline, with every decision and every comment in one place. The paperwork that used to slow things down now moves on its own.
