Vexp: Dependency Graph Context for AI Agents

If an entrepreneur "adopted" this product today, here's the realistic math.

Fermi summary

Get 300 paying devs at $20/mo and you hit $72k ARR - but you're one Cursor changelog entry away from zero, so the 14% success odds are the real story.

Market size (TAM)

$48.0M

~200k developers actively using AI coding agents on complex multi-module codebases who would pay for context enrichment × $20/mo avg

Year-1 ARR range

$18k - $240k

midpoint $72k

Gross margin

82%

Investment to production

$22k

Dev $10k: billing, auth hardening, API stability, edge-case fixes for monorepos. Marketing $8k: HN launch, Cursor/Windsurf Discord presence,

Probability of success

14%

P(reaching mid case in 12 months)

Expected take-home Y1

$-13500

probability-weighted, after investment

Go-to-market motion

Product-led growth: free tier on Cursor marketplace / GitHub → organic HN/Reddit/X dev community → paid conversion at ~$20/mo individual or $80/mo team, targeting 300 paid users by month 12.

Key risks

Cursor, Windsurf, or GitHub Copilot ships native dependency-graph context as a built-in feature - likely within 12 months given the obvious demand - commoditizing the entire product overnight
Context windows keep expanding (200k+) and models get better at inferring structure from partial code, shrinking the pain point that justifies the purchase
Developers treat this as a 'script it in an afternoon' problem - tree-sitter, ast-grep, and .cursorrules already let technically strong devs roll their own for free, limiting willingness to pay

Generated by the Wishdeal Factory financial-analysis agent. Numbers are honest Fermi estimates, not guarantees. Real outcomes depend on the operator. The studio is bullish on the engineering quality, agnostic on the business outcome.