Fabre AI — Investor & Partner Pitch Document

CS2 Tactical Intelligence Platform | AI-powered demo analysis for professional esports

Executive Summary

Fabre AI is a B2B SaaS platform that transforms raw CS2 match recordings (.dem files) into actionable tactical intelligence. Using a proprietary tick-level parsing engine combined with AI-driven pattern recognition, we deliver the kind of analysis that previously required a full-time analyst team — in seconds, at a fraction of the cost.

We target professional CS2 teams, esports coaches, and competitive communities through a subscription model, with pricing tiers ranging from free exploration to premium counter-strategy suites.

The Problem

Counter-Strike is a data-rich game with a data-poor analysis ecosystem

Every CS2 match generates a binary demo file containing millions of data points — player positions, utility usage, economy decisions, and movement patterns across every single game tick (64-128 per second). Yet the current tools available to teams are either:

Tool	Limitation
Manual VoD review	Time-intensive, subjective, no data aggregation
HLTV / Leetify	Surface-level stats, no deep tactical pattern extraction
Proprietary org tools	Exist only at top-5 org level; cost $50K–$200K/year to build in-house
Raw demo parsers	Developer-only, require significant engineering effort to extract insights

The result: 95% of professional and semi-professional teams make tactical decisions based on intuition and manual film study rather than systematic data analysis.

The Solution

Fabre AI: From .dem file to tactical report in under 60 seconds

Upload .dem file
      |
      v
[Tick-Level Parser] — 64–128 snapshots/sec, every player event
      |
      v
[Tactical Analyzer] — Rush detection, lurk patterns, utility efficiency,
                       site preference, economy decisions
      |
      v
[AI Insight Engine] — Gemini / GPT-4o synthesis of patterns into
                       human-readable tactical reports
      |
      v
[Structured Report] — JSON + Dashboard visualization

Coaches receive ranked key findings, confidence scores, and specific counter-strategy recommendations — not raw data dumps.

Product: What We Deliver

5 Distinct Analysis Modules

Module	What It Detects	Who Uses It
Utility Tracking	Smoke, flash, HE, molotov usage by player/round/site	All tiers
Position Analysis	Location heatmaps, weapon-position correlation, map control dominance	All tiers
Round Intelligence	Rush detection (4+ players), lurk patterns, smoke gap kills, pivot plays	Basic+
AI Tactical Report	Player tendency profiles, team execute patterns, turning point analysis	Pro+
Counter-Strategy	Opponent weakness mapping, multi-match aggregation, AI-generated counters	Premium

Sample Output — AI Tactical Report

json

{
  "type": "team_tactics",
  "key_findings": [
    "Aurora executes B-site on 73% of CT-side rounds using a 3-man default split",
    "biguzera averages 2.4 smokes per round — highest utility efficiency in dataset",
    "PAIN consistently trades AWP for rifle in eco rounds (82% of buy decisions)"
  ],
  "counter_recommendations": [
    "Stack 3 on B early — Aurora rarely re-takes from CT mid",
    "Force-buy round 8: PAIN buys down when behind by 3+ rounds"
  ],
  "confidence": 87
}

Market Opportunity

CS2 is the world's most-watched PC esport

Metric	Value
Active CS2 players (monthly)	32 million
Professional teams globally	~2,000 (Tier 1–3)
Semi-pro / academy teams	~15,000
Major tournament prize pool (2024)	$20M+
Global esports market (2025)	$2.1B
Analytics software segment CAGR	18.4%

Addressable Market

TAM (Total): $420M — All esports analytics across all titles
SAM (Serviceable): $85M — CS2-specific performance analytics
SOM (Obtainable — Year 1): $2.1M — 300 Pro/Premium subscribers + 1,200 Basic

Business Model

Subscription Tiers

Tier	Price	Monthly Match Quota	Key Features
Free	$0	3 matches	Utility + Position
Basic	$29/mo	20 matches	+ Round detail reports
Pro	$79/mo	50 matches	+ AI tactical analysis
Premium	$199/mo	Unlimited	+ Counter-strategy + Multi-match aggregation

Revenue Drivers

Primary: Recurring SaaS subscriptions
Secondary: Team/org enterprise contracts (custom integrations, private API access)
Tertiary: Tournament organizer licenses (post-match public analysis packages)

Unit Economics (Projected — Year 1)

Metric	Value
Target MRR (Month 12)	$45,000
Avg. Revenue per User	$68/mo
Estimated CAC	$120
Payback Period	~1.8 months
Gross Margin (excl. AI compute)	~78%

Competitive Landscape

Competitor	Strength	Our Edge
Leetify	Large user base, good UX	We go tick-level; Leetify is summary stats only
FACEIT Analytics	Platform integration	Platform-agnostic; works on all official matches
SportsCode / Catapult	Enterprise sports analytics	CS2-native; Catapult has no demo parser
In-house org tools	Custom-fit	100x cheaper; no engineering team required
HLTV Stats	Brand recognition	We provide tactical insights, not leaderboard stats

Our Defensible Advantages

Tick-level resolution — We parse every game frame (64–128/sec), not sampled snapshots. This enables rush detection, execute timing analysis, and lurk pattern recognition that summary-stat tools cannot replicate.
AI synthesis layer — Raw tactical data is only useful when interpreted. Our prompt templates, trained on CS2-specific tactical vocabulary, convert patterns into coach-ready language.
Multi-match aggregation — Single-game analysis misses meta-patterns. Our Premium tier builds opponent profiles across entire tournament runs.
Cost-efficient AI pipeline — MatchDataCompressor reduces token usage by ~60% before LLM calls. A full match analysis costs ~$0.001 in compute.

Technology Stack

Backend:      TypeScript 5.3 / Node.js 20 / Express.js
Parser:       @laihoe/demoparser2 (Rust-based, handles 500MB demos)
Database:     PostgreSQL 15 on Google Cloud SQL
AI:           Google Gemini 2.0 Flash + OpenAI GPT-4o (provider-agnostic)
Infrastructure: Google Cloud Run (auto-scaling microservices)
Storage:      Google Cloud Storage (demo files)
Queue:        Google Cloud Tasks (async processing)
Auth:         JWT + Stripe / Iyzico payment integration
CI/CD:        Cloud Build → Container Registry → Cloud Run

Architecture: Fully Cloud-Native

Zero-idle cost: Cloud Run scales to zero between jobs
Horizontal scale: Each analysis job is a stateless container — handle 1 or 1,000 concurrent uploads without re-architecture
Compliance-ready: Demo files stored in regional GCS buckets (GDPR zone isolation)

Traction & Validation

Technical Milestones (Completed)

Phase	Status	Description
Tick-Level Parser	Complete	64-tick parsing, all player events captured
Utility Tracking	Complete	Smoke/flash/HE/molotov by player, round, team
Position Analysis	Complete	Heatmaps, weapon correlation, map control
AI Pipeline	Complete	5 prompt templates, Gemini + OpenAI support
Team Comparator	Complete	Head-to-head tactical diffs
Multi-Match Profiling	Complete	Opponent profile aggregation
REST API	Complete	JWT auth, rate limiting, job queue
Cloud Run Deployment	Complete	Production microservice architecture
Frontend Prototype	In Progress	Radar visualization, live demo available

Live Demo

Real match analyzed: Aurora vs PAIN — de_nuke (ESL Pro League format)

24 rounds parsed at tick-level
Utility efficiency, position heatmaps, and AI tactical report generated
Interactive radar map with player position overlays (multi-floor support)

Go-to-Market Strategy

Phase 1 — Community (Months 1–3)

Launch free tier on FACEIT / Reddit CS2 communities
Partner with 3–5 Tier 2/3 tournament organizers for free analysis packages
Content: "We analyzed [major match] so you don't have to" posts

Phase 2 — Semi-Pro Conversion (Months 3–9)

Target IEM/ESL Challenger-level teams ($29–$79/mo)
Offer 30-day free Pro trials to teams entering open qualifiers
Coach referral program: 20% recurring commission

Phase 3 — Enterprise / Org Contracts (Month 9+)

Outbound to Tier 1 org analytics directors
Custom integration + private API + white-label reports
Pricing: $2,000–$8,000/mo per org

Team

Role	Background
Founder / Tech Lead	Full-stack + systems engineering; built the entire parser + AI pipeline from scratch
Open — Growth Lead	Esports industry experience preferred
Open — Frontend Engineer	React + data visualization

Why Now

CS2 replaced CS:GO in 2023 — the ecosystem is rebuilding. Early movers in analytics tooling will own the category.
AI inference costs dropped 10x in 18 months — what cost $5/match in GPT-4 now costs $0.001 with Gemini Flash.
Esports orgs are professionalizing — investment in coaching staff and performance analytics is at an all-time high.
Demo format is stable — CS2's .dem format is well-documented and our parser handles live tournament demos day-of-release.

Contact

Fabre AI Email: hello@fabre.ai GitHub: github.com/fabre-ai Demo: [Live prototype available on request]

This document contains forward-looking statements. Financial projections are estimates based on market research and internal modeling.

Fabre AI — Investor & Partner Pitch Document ​

Executive Summary ​

The Problem ​

Counter-Strike is a data-rich game with a data-poor analysis ecosystem ​

The Solution ​

Fabre AI: From .dem file to tactical report in under 60 seconds ​

Product: What We Deliver ​

5 Distinct Analysis Modules ​

Sample Output — AI Tactical Report ​

Market Opportunity ​

CS2 is the world's most-watched PC esport ​

Addressable Market ​

Business Model ​

Subscription Tiers ​

Revenue Drivers ​

Unit Economics (Projected — Year 1) ​

Competitive Landscape ​

Our Defensible Advantages ​

Technology Stack ​

Architecture: Fully Cloud-Native ​

Traction & Validation ​

Technical Milestones (Completed) ​

Live Demo ​

Go-to-Market Strategy ​

Phase 1 — Community (Months 1–3) ​

Phase 2 — Semi-Pro Conversion (Months 3–9) ​

Phase 3 — Enterprise / Org Contracts (Month 9+) ​

Team ​

Why Now ​

Contact ​