Private Alpha · Season 1

Whoever ships
first wins.

Real-time 1v1 coding battles. Get matched by ELO, solve under a 15-minute clock, climb the global ladder.

View Leaderboard →
solution.pyopponent
⏱ 09:42+140
1
2
3
4
5
6
7
8
9
10
# Two Sum — solve before your opponent does
 
def twoSum(nums: list[int], target: int) → list[int]:
    seen = {}
    for i, n in enumerate(nums):
        if target - n in seen:
            return [seen[target - n], i]
        seen[n] = i

Matches Played1,452,030
Active Duels1,205
Developers50,000+
Avg Latency24ms

How it works

Three steps to dominance.

01
Match

Get paired with a developer at your exact ELO. The system matches you in seconds — no lobby, no waiting.

02
Solve

Race on the same problem under a shared 15-minute clock. Write clean code — not just fast code.

03
Climb

First correct submission wins the ELO delta. Ties go to the AI tiebreaker judging on complexity and code quality.


Rank System

Where do you stand?

Bronze
800–1199 ELO
Silver
1200–1499 ELO
Gold
1500–1799 ELO
Diamond
1800–2099 ELO
Master
2100+ ELO

ELO Rating

A fair fight, every time.

CodeDuel uses chess-style ELO ratings. Win against a higher-rated opponent and you gain more. Lose to a lower-rated one and you lose more. The system self-balances — your rank reflects reality.

Expected win vs equal opponent50%
Rating points per win (K=32)~+16
Max swing per match±32
PU
purahan
18421860 ELO
+18
GH
ghost_coder
18191801 ELO
-18
Two Sum · purahan submitted O(n) hash map solution in 4:21. ghost_coder's O(n²) approach ran 230× slower on the hidden test case.

Live Activity updated 2m ago
PlayersProblemDifficultyResult
ND
CN
NeoDev vs CodeNinja
Two Sum OptimizationeasyWin  +15 ELO
NP
BN
NullPointer vs ByteNinja
LRU Cache ImplementationhardLoss  −12 ELO
RK
AD
RecursionKing vs AlexDev
Word Search IIhardWin  +24 ELO
AK
SX
AlgoKing vs SyntaxError
Trapping Rain WatermediumWin  +18 ELO
PU
BH
purahan vs ByteHunter
Course SchedulemediumWin  +21 ELO

AI Tiebreaker

First to submit isn't always the winner.

When two players submit within the same window, a deterministic layer measures real runtime and complexity — then an AI narrates exactly why one solution won.

  • Runtime and memory measured from actual execution, not self-reported
  • Complexity class inferred from benchmark curves — O(n) vs O(n²)
  • AI explains the verdict in plain English, can't override the score
Tiebreaker VerdictAI Judge
Both players submitted a correct solution within 1.2 seconds of each other. purahan's solution runs in O(n) using a hash map, while ghost_coder's nested loop approach is O(n²). On the largest hidden test case (n=10⁵), the difference is 42ms vs 9,800ms.
Winnerpurahan+22 ELO

Social Play

Settle the debate.

Add your coworkers, classmates, or rivals. Challenge them to private duels to definitively prove who the better developer is. Track your head-to-head records and climb your private leaderboards.

  • Global friend search via GitHub handle
  • Direct challenge notifications
  • Private match arenas without ELO stakes
ghost_coder
Online
CHALLENGE
alex_dev
In a match
SPECTATE

Community

Built by developers, for developers.

CodeDuel is open source and entirely community driven. Have a feature request, found a bug, or want to contribute new algorithmic challenges?

GitHub RepositoryContact Us

Ready to compete?

Prove your worth in real time.

Join 50,000+ developers. Your first match is one click away.

Built with
AWS DynamoDB
Vercel
Next.js
Piston API
Gemini AI
Aurora PostgreSQL