ForeSight

Inspiration

Venture capital firms invest in companies with enormous growth potential, but understanding how those companies are actually performing after investing remains a challenge. Financial statements, founder updates, operational KPIs, and board reports are often scattered across spreadsheets, emails, and documents.

Most portfolio monitoring tools help investors understand what has already happened. We wanted to take that a step further: What if investors could identify emerging problems before they become critical and understand the potential financial consequences of acting on them?

That question inspired ForeSight.

What it does

ForeSight is an AI-powered portfolio intelligence platform that helps venture capital firms move from reactive reporting to proactive decision-making.

Our platform brings together company financials, operating metrics, management forecasts, and qualitative updates to give investors a comprehensive view of their portfolio.

ForeSight is designed to:

Detect emerging risks: Identify deteriorating financial and operational trends, including weakening customer retention, rising cash burn, missed forecasts, and liquidity concerns.
Provide meaningful dashboards: Give investors company-specific insights using relevant industry KPIs, with customizable metrics and reporting views.
Support custom financial formulas: Accommodate both founder-defined and investor-defined metrics.
Analyze business context: Connect quantitative performance with founder updates, strategic milestones, and previous management decisions.
Evaluate potential decisions: Use scenario modeling and sensitivity analysis to explore how changes in hiring, spending, growth, or operating assumptions could affect future financial performance.
Generate actionable insights: Help investors identify what deserves attention, why it matters, and which questions to ask management.

Rather than simply displaying another red warning indicator, ForeSight aims to explain the underlying problem and help investors evaluate potential responses.

How we built it

We designed ForeSight around a structured financial database, deterministic analytical rules, and an AI-assisted intelligence layer.

We created a synthetic portfolio of 25 venture-backed companies, with financial histories, operating KPIs, forecasts, and company-specific scenarios designed to represent realistic business challenges.

Our financial intelligence layer validates company information and applies deterministic rules to identify issues without relying on an LLM to perform financial calculations.

For example, we implemented a retention deterioration rule that analyzes six consecutive months of Net Revenue Retention (NRR) and Gross Revenue Retention (GRR). Instead of only flagging companies after their retention falls below a threshold, it can identify sustained deterioration and raise an earlier warning.

Our approach separates responsibilities: deterministic calculations establish what the numbers show, while AI helps interpret business context, investigate risks, and communicate findings.

We also structured the platform around reusable financial capabilities and industry-specific metrics, creating a foundation for analyzing different business models rather than restricting the product to SaaS companies.

Challenges we ran into

One of our biggest challenges was creating realistic financial data. Generating random numbers would have produced inconsistencies between revenue, customers, operating expenses, cash balances, and forecasts.

We needed the data to reflect actual business relationships so that emerging risks would be financially meaningful.

Another challenge was minimizing false positives. A temporary seasonal decline shouldn't automatically trigger the same warning as sustained operating deterioration.

We also encountered technical challenges integrating our financial rules with PostgreSQL. An SQL parameterization issue initially caused the alert pipeline to fail. After identifying and correcting the issue, we successfully reran the database checks and automated tests.

Finally, we had to balance AI flexibility with financial reliability. We wanted AI to provide valuable explanations without fabricating numbers, assumptions, or unsupported conclusions.

Accomplishments that we're proud of

We're particularly proud of building a foundation that prioritizes financial accuracy, explainability, and proactive detection.

Our validated financial-rule pipeline generated:

1,732 verification checks
1,339 verified claims
52 alerts: 17 critical, 33 warnings, and 2 informational
10 passing automated tests

Our retention deterioration rule successfully identified emerging concerns in companies with weakening retention while avoiding unnecessary alerts for healthy businesses, seasonal fluctuations, and isolated customer churn events.

We also developed a cross-industry data architecture designed to support enterprise SaaS and AI, fintech, healthcare and biotech, deep technology, and consumer marketplaces.

Most importantly, we've established a foundation for connecting financial signals with the business decisions investors actually need to make.

What we learned

We learned that identifying a problem is not the same as understanding it.

A company can experience strong revenue growth while its customer retention weakens, acquisition costs rise, or cash runway deteriorates.

A useful portfolio intelligence system needs to examine relationships between metrics, understand the company's business model, and distinguish temporary fluctuations from structural problems.

We also learned that AI and traditional financial modeling work best together. AI is valuable for understanding qualitative updates, retrieving relevant information, and explaining findings. Deterministic models are better suited for accounting reconciliations, KPI calculations, and financial scenario analysis.

Above all, we learned that trustworthy intelligence requires reliable data, transparent assumptions, and traceable evidence.

What's next for ForeSight

Our vision is to make ForeSight a comprehensive decision-intelligence platform for venture capital firms.

Next, we want to expand our industry-specific analytical capabilities, deepen our AI agent orchestration, and connect emerging risks directly to financial scenario simulations.

We also plan to make dashboards and financial formulas more customizable, integrate with existing accounting and portfolio management systems, and develop a feedback loop that compares previous recommendations with actual business outcomes.

Ultimately, our goal is simple:

ForeSight shouldn't just tell investors which companies need attention. It should help them understand why, what could happen next, and which decisions are worth evaluating before it's too late.