Method

What is .Déjà?

.Déjà's business assessment framework is designed with input from investors and business operators, drawing on their combined experience to assess tech companies at different points in history. Despite rapid advances in large language models (LLMs) and other artificial intelligence (AI) systems, comparable historical cases for assessing tech companies remain scarce, especially among successful tech companies. For that reason, .Déjà's prediction research compares TabPFN-v2, a pretrained foundation model for tabular data, with a valuation baseline and XGBoost. As AI becomes a new layer of infrastructure, some of the patterns seen in dot-com companies during the internet bubble may reappear in today’s AI-native companies.

How do agents participate?

For each company and target date, human analysts work with AI agents to reconstruct what could reasonably have been known at the time. The website’s company discovery flow also uses these agents: when a company is not yet in the catalogue, they help identify it from a name or website before evidence gathering and assessment. Analysts apply the framework, while agents handle repeatable tasks such as collecting evidence, checking dates, and making first-pass assessments of individual dimensions.

What is the goal?

The goal is to present an overall profile of a company at a specific point in history, rather than rely on an LLM to generate the assessment directly. Data fitting takes place after evidence collection and scoring.

For deeper company insights, join the wait list. For questions about the assessment methods or research dataset, contact the developer.