How FIFA pre-sets the 2026 World Cup bracket before third-place teams are known

·1 views

The 2026 FIFA World Cup features 48 teams across 12 groups, with the top two from each group advancing automatically to make 24 qualifiers. To complete a 32-team knockout bracket, the eight best third-placed finishers from across all 12 groups are also admitted, but their identities are unknown until the group stage ends. Despite this uncertainty, the bracket structure — determining who plays whom and where — is fixed well in advance of any matches being played. FIFA achieves this by pre-mapping all possible combinations of which eight groups produce the qualifying third-place teams, assigning each combination a predetermined set of matchups. A key constraint in the draw is that no third-place team can be drawn against the group winner from their own group, since the two sides already met during the group stage.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Why AI Agents Fail in Production: The Case for Idempotent Design

A technical analysis published on DEV Community argues that most production AI agent failures stem not from flawed reasoning but from unreliable network conditions common to all distributed systems. Write-capable agents — those that can send emails, charge payments, or update databases — are vulnerable to duplicate actions when retries follow timed-out requests that already succeeded server-side. The author illustrates this with a double-invoice scenario where a perfectly functioning model retries a call it never received confirmation for, resulting in two real-world transactions. The proposed fix borrows from payments infrastructure: attaching idempotency keys to every side-effecting action, so that retried calls return the stored result of the original rather than triggering a second operation. For agents lacking human click events, the key is derived deterministically from the tool name and its parameters, ensuring the same logical intent always maps to the same key across retries and restarts.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

How AI Can Help Engineers Diagnose Server Log Errors Without Replacing Human Judgment

A DevOps engineer describes using AI language models to analyze large volumes of server logs during incidents, such as tracking down why a customer instance fails to receive a floating IP at 2 a.m. The core argument is that AI is valuable for pattern-matching and correlating across tens of thousands of log lines from multiple services, translating technical jargon into plain English. However, the author stresses that the model should only surface ranked hypotheses and verification commands — never autonomously apply fixes — keeping the engineer as the final decision-maker. Before sharing any logs with an AI tool, the author recommends running an automated redaction pass to strip out tokens, passwords, private IPs, and other sensitive data. Building redaction directly into the log-pull command, rather than treating it as a separate manual step, is highlighted as a critical operational habit.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Structured Processes Beat New AI Tools, Says Engineer Behind PathPilot

A software engineer argues that inconsistent internal processes, not a lack of AI, are the core operational problem facing growing businesses. Employees handling the same situation differently signals a process gap, not a technology gap. Existing documentation tools like Notion or Confluence often go unused under real work pressure because searching lengthy documents is impractical. To address this, the engineer built PathPilot, a visual workflow tool that converts standard operating procedures into interactive decision flows. The product aims to make processes easier to follow in the moment, and the author contends that well-defined workflows also make AI integrations more effective.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer Builds Three Token-2022 Mints on Solana Featuring Fees, Yield, and Transfer Locks

A developer participating in the 100 Days Of Solana challenge built three distinct token mints using Solana's Token-2022 program and its extension system. The first mint enabled transfer fees set at 1% per transaction, enforced at the protocol level for use cases like DAO treasuries and creator royalties. The second mint used an interest-bearing extension that displays a computed yield to users without actually minting new tokens or changing the total supply. The third mint was configured as non-transferable, with the Token-2022 program itself blocking any transfer attempts on-chain, making it suitable for identity credentials and membership passes. The project demonstrated that Token-2022 extensions allow developers to add advanced token behaviors without building custom smart contracts from scratch.

0 comments Read more at DEV Community