
Desia Team
PRODUCT
April 2, 2026
Vesper Infrastructure Partners specializes in mid-cap value-add infrastructure across Europe. With over €1 billion in assets under management, the Vesper Infrastructure Next Generation Fund I focuses on value-add infrastructure assets where accuracy, consistency, and defensible reasoning are mandatory for evaluating potential target companies.
Traditional screening is a typical manual bottleneck. Evaluating a single target against Vesper’s proprietary investment checklist previously required a full day of intensive work from a senior analyst. The process involves a rigorous data extraction phase where information is synthesized from disparate sources:
Analysts must cross-reference this data against multiple regulatory frameworks, including EU directives, and internal classification frameworks. The final step is to assign a score that is robust and consistent with the firm's historical judgment.
Maintaining this level of rigor manually requires not only a substantial amount of analysts’ time but also considerable senior review and supervision, which materially limits the volume of opportunities a lean team can process.
Vesper’s objective was to build an AI scoring engine to maximise internal resources’ productivity by automating this extraction and scoring logic while providing robust and auditable justification for every point awarded.

Working with the Vesper team, we builta dedicated screening workflow on Desia that translates their investment screening model into an automated process. The system takes a company name as input and produces a fully scored investment assessment in less than 15 minutes.
The workflow is engineered to function within Vesper’s operational parameters:
By integrating internal documents, web sources, and financial databases simultaneously, the platform performs the complex synthesis that previously required hours of manual research.
Building this robust AI workflow was a collaborative engineering process focused on capturing the nuances of Vesper's investment screening philosophy.
The build process involved several intensive stages:
By encoding these specific steps, the workflow ensures that every score is a defensible conclusion grounded in Vesper's established methodology, within their own internal structure.
In the first six months of this partnership, the transition to an automated workflow has delivered measurable operational improvements:
Together, we will continue to refine the current model, expand coverage, and ensure the system remains aligned with evolving regulatory frameworks. This partnership focuses on building a continuous pipeline of tools that encode Vesper’s specific analytical engine.
For Desia, this reflects our approach to building specialized tools that capture how expert teams think. We look forward to developing additional workflows as Vesper continues to scale its investment strategy across Europe.
Our large-scale natural language processing allows you to search, generate documents,and uncover answers to complex questions in minutes.
In Q3 2023, Company XYZ faced a major data breach, exposing customer information. The incident led to a 30% churn rate among enterprise clients over the next two months. Consequently, annual recurring revenue dropped from $30M to $23M by year-end 2023.