Building OpponentIQ: Automating Esports Scouting with GRID API

Dhani Dzulkarnain — Fri, 23 Jan 2026 08:08:55 +0000

The Problem

In competitive esports, preparation is everything. Professional teams spend hours analyzing opponents through VOD reviews, manually tracking statistics, and compiling notes from scattered data sources. What if we could automate this entire process using real competitive data?

That's exactly what I set out to build for the Sky's the Limit - Cloud9 x JetBrains Hackathon.

What is OpponentIQ?

OpponentIQ is an AI-powered scouting report generator that automatically analyzes any League of Legends or VALORANT team using official match data from GRID API. In seconds, coaches and players get a comprehensive breakdown including:

Team strategies: Win rates, composition patterns, and macro tendencies
Player analysis: Individual stats, champion/agent pools, and KDA metrics
Actionable insights: Data-driven recommendations like "Neutralize [Player] - target bans recommended"
Export capabilities: Professional PDF reports and JSON data

Technical Architecture

Backend: Python + FastAPI

I chose FastAPI for its speed and automatic API documentation. The backend handles three main responsibilities:

GRID API Integration - Fetching real match data via GraphQL
Data Processing - Analyzing patterns and calculating statistics
Report Generation - Transforming raw data into actionable insights

The GRID API Challenge

GRID's esports data is incredibly comprehensive but split across two endpoints:

Central Data API: Team metadata and series IDs
Series State API: Detailed game statistics

I implemented a two-stage retrieval system that fetches series information first, then queries detailed stats for each match. This required careful GraphQL query optimization to keep response times under 5 seconds even when analyzing 10+ matches.

# Simplified example of the two-stage approach
def fetch_team_matches(team_name, num_matches):
    # Stage 1: Get series IDs from Central Data
    series_ids = central_data_query(team_name, num_matches)

    # Stage 2: Get detailed stats from Series State
    matches_data = []
    for series_id in series_ids:
        match_data = series_state_query(series_id)
        matches_data.append(match_data)

    return matches_data

Data Analysis Engine

The scouting_analyzer.py module processes match data to generate insights:

For League of Legends:

Infers player roles (not provided by API)
Identifies champion pools and ban priorities
Calculates KDA ratios and performance trends
Detects team composition patterns

For VALORANT:

Tracks agent selections and map preferences
Calculates Average Combat Score (ACS)
Analyzes headshot percentages
Identifies team composition meta

Frontend: Clean and Responsive

I built the UI with HTML5, Bootstrap 5, and vanilla JavaScript. Key features:

Galaxy-themed animation: A stunning opening curtain reveal
Interactive charts: Chart.js visualizations for performance trends
Real-time loading: Progress indicators with status updates
Export functionality: PDF generation using jsPDF

Development Experience with PyCharm

Using JetBrains PyCharm throughout this project was a game-changer:

Intelligent code completion saved hours during FastAPI development
Integrated debugging helped solve the role inference logic quickly
GraphQL query formatting made API integration cleaner
Git integration streamlined version control

Key Challenges

1. Data Transparency

When GRID API doesn't provide certain data (like specific VALORANT map names in some responses), I had a choice: fabricate placeholder data or be transparent.

I chose transparency. OpponentIQ displays "Unknown" when data isn't available, maintaining credibility over aesthetics.

2. Role Inference for League of Legends

Player roles aren't explicitly provided by GRID's LoL API. I implemented logic to infer roles based on:

Champion selections (e.g., Jarvan IV → likely Jungle)
Game positions and farm patterns
Historical role patterns

3. Generating Actionable Insights

Raw statistics are useful, but coaches need actionable strategies. I built algorithms that:

Identify star players (high KDA) and recommend target bans
Detect weak performers and suggest lane exploitation
Analyze team form and recommend aggression levels
Parse ban priorities to suggest counter-picks

Example insight generated:
"Neutralize Keria"Keria (Support) has 4.64 KDA - target bans on Milio recommended

Results

OpponentIQ successfully generates professional scouting reports with:

✅ 100% live GRID API integration (zero mock data)
✅ Multi-game support (League of Legends + VALORANT)
✅ PDF export for team meetings
✅ Historical tracking for opponent evolution
✅ Sub-5-second response times for 10-match analysis

What I Learned

GraphQL optimization is critical for performance with nested queries
Data transparency builds more trust than fake placeholders
User experience matters - loading animations and progress bars make waiting feel shorter
Production-ready means handling edge cases gracefully

Try It Yourself

The project is fully open source:

GitHub: https://github.com/DhaniDzulkarnain1/Cloud9-Hackathon

Demo Video: https://youtu.be/QvMQ0h0xhCM

Quick Start:

# Clone the repository
git clone https://github.com/DhaniDzulkarnain1/Cloud9-Hackathon.git

# Install dependencies
pip install -r requirements.txt

# Run the backend
cd backend
python3 main.py

# Open frontend/index.html in your browser

What's Next?

I'm planning to expand OpponentIQ with:

More games: CS:GO, Dota 2, Rocket League
Win prediction: Machine learning models for match outcomes
Team comparison: Side-by-side opponent analysis
Live integration: Real-time scouting during tournaments

Final Thoughts

Building OpponentIQ taught me that the best tools solve real problems. Esports teams need efficient opponent analysis, and by leveraging GRID's official data API, I created something that could genuinely help coaches and players prepare for matches.

If you're interested in esports analytics, data visualization, or just want to see how to integrate the GRID API into your projects, check out the code on GitHub!

Built for: Sky's the Limit - Cloud9 x JetBrains Hackathon 2025
Category: Category 2 - GRID API Integration
Tech Stack: Python, FastAPI, GRID API, Bootstrap 5, Chart.js, JetBrains PyCharm

Connecting database with API

Dhani Dzulkarnain — Tue, 19 Nov 2024 09:46:28 +0000

can anyone help me, so I made a mobile platform and the two websites are integrated with each other, then this website functions as a collector and mobile as a fisherman, the relationship between these two platforms is to record fish catches, and for this mobile I have created a frontend display using Jetpack Compose, then for the backend, how can I connect via API, for the database created using MySQL.

DEV Community: Dhani Dzulkarnain