ThesisModelVenturesTeamInsightsBuild With Us
All Case Studies

StockSense
by Diversss

Wealth manager-grade stock analysis — for every retail investor, in seconds. Powered by Claude API and a proprietary multi-factor scoring framework built by Diversss.

4Scoring Pillars
25%ROCE/PE Quality Weight
SecondsPer Full Analysis
Claude APIIntelligence Layer

There are roughly 160 million retail investors in India. Almost none of them have access to the kind of stock analysis that actually informs good investment decisions — the kind that looks at return quality, growth trajectory, technical positioning, and institutional flow simultaneously and synthesises them into a clear, actionable verdict.

That quality of analysis has always existed. It lives inside the research desks of wealth management firms and institutional brokerages. It is produced by analysts who have spent years building frameworks for evaluating stocks across multiple dimensions. And it is reserved, almost entirely, for clients with significant assets under management.

StockSense was built to change that.

The Problem with Retail Stock Research

Most retail investors make decisions based on incomplete information: a news headline, a tip from a friend, a price chart, or a YouTube video. Not because they don't care — but because the good analytical frameworks are either inaccessible, incomprehensible, or prohibitively expensive to access.

The result is predictable. According to SEBI data, ~73% of retail investors cite research gap as the primary reason for poor portfolio outcomes. Over 10 million new demat accounts were opened in FY24 alone — most of them entering the market without a reliable analytical framework.

"Diversss built the framework. Claude provides the reasoning. Together they produce verdicts that used to require an analyst and a week of research."

The Four-Pillar Scoring Framework

StockSense is built on a proprietary scoring framework developed by the Diversss team — four core pillars that assess a stock from the angles that actually matter for long-term returns.

25%
ROCE / PE Quality
Return on capital employed and price-earnings analysis — measuring whether a company earns well on the capital it deploys, and whether you're paying a fair price to own it.
15%
Revenue & Profit Growth
Multi-year revenue and profit growth trajectory — filtering for businesses that are genuinely expanding, not just reporting one-off improvements.
Variable
Technical Composite
Price momentum, volume patterns, and relative strength — identifying whether the stock is technically positioned to move in the direction the fundamentals suggest.
10%
FII / DII & MF Flow
Institutional and mutual fund flow data — tracking where smart money is moving, as a proxy for informed conviction in the stock.

How Claude API Powers the Intelligence Layer

The scoring framework provides the structured inputs — quantitative signals across the four pillars, weighted and normalised. Claude API provides the intelligence layer that converts those signals into analysis a retail investor can actually use.

This is a critical architectural distinction. Claude is not a chatbot in this system. It is a structured reasoning engine. It receives the scored data, the framework weights, and the specific stock context — and it returns a verdict structured around the four output categories that every investor actually needs:

Strengths
What the stock has going for it right now
Concerns
The risks and weak signals worth watching
Hope Factors
Catalysts that could change the picture ahead
Next Action
A clear, framework-backed recommendation

Beyond the core scoring, StockSense includes a Live Signal layer that correlates news, earnings announcements, policy changes, sector moves, and FII/DII flow events to specific holdings in a user's portfolio — surfacing what's relevant, not just what's happening.

The Correlation Engine analyses portfolio-level dynamics — sector overlap, risk concentration, and true diversification — ensuring users understand not just individual stocks, but how their holdings interact as a system.

What This Actually Delivers

A senior portfolio manager at a wealth management firm would take hours — sometimes days — to produce the quality of analysis StockSense generates in seconds. They would charge for the privilege, typically requiring ₹40L+ in minimum AUM before you qualify for that level of service.

StockSense makes that standard of analysis available to every retail investor who downloads Diversss. Not a simplified version. Not a chatbot that tells you to "do your own research." A structured, multi-dimensional, framework-driven verdict — in seconds.

That is what happens when you separate the scoring framework from the reasoning engine, and build each one properly.

Work With TEN Labs

Building in fintech, investing, or wealth tech?

If you have a domain expertise framework and want to wire Claude API to it to create genuinely intelligent products, we've done exactly this. Let's talk about what's possible for your use case.

Start a Conversation