Automation / DevOps / IaC

HashiCorp's Terraform continues to dominate GitHub trending lists, serving as the primary tool for safely creating and versioning infrastructure. It codifies cloud and on-prem APIs into declarative configuration files that teams can share and treat as code. The platform supports both popular service providers and custom in-house solutions through its source-available model.

• Declarative config files enable safe, predictable infrastructure changes across teams.
• Supports existing providers and custom in-house infrastructure solutions.
• Version control and code review workflows are integral to the IaC process.
• Certification and learning resources available via HashiCorp's official platforms.

Datadog engineer Arnold Wakim details a production migration where AI tools Claude and Cursor helped overcome storage backend limitations. The team leveraged these models to refactor critical infrastructure, resulting in significant performance improvements. The process highlights both effective strategies and pitfalls encountered during this AI-assisted engineering effort.

• AI pair programmers can assist with complex refactoring in production environments
• Storage backend limits may require innovative architectural solutions
• Performance gains justify the effort of AI-assisted migration strategies
• Reviewing what didn't work is as valuable as documenting successes

Percona Operator for MySQL 1.2.0 introduces cross-site replication to support multi-region deployments. The update also adds support for encrypted backups stored in object storage and automatic storage scaling. These features address common operational gaps for fleets expanding beyond a single cluster.

• Enables warm replica clusters in second data centers for improved disaster recovery.
• Supports encrypted backups in object storage to meet auditor requirements.
• Adds automatic storage scaling to reduce manual intervention during growth.

Bryan Oliver presents techniques for applying chaos engineering to complex AI infrastructure, focusing on large GPU clusters. The talk addresses handling intricate topologies, RDMA network protocols, and NUMA misalignments through practical fault injection. The goal is to maximize hardware efficiency and establish robust observability loops for multi-million dollar investments.

• Apply chaos engineering to validate stability in large-scale GPU clusters.
• Address specific infrastructure challenges like RDMA and NUMA misalignments.
• Implement seven practical fault-injection strategies for robust testing.
• Build observability loops to ensure efficiency and reliability of AI hardware.

AWS published a guide to building an automated pipeline that extracts PostgreSQL audit logs from CloudWatch Logs. The solution converts these entries into structured CSV format in near real time and stores them in Amazon S3 for long-term analysis.

• Eliminates manual log retrieval by automating extraction from CloudWatch Logs to S3.
• Converts unstructured log data into structured CSV for easier parsing and analysis.
• Enables near real-time processing of audit events for timely security monitoring.
• Leverages S3 for durable, long-term storage of compliance and audit trails.

GitHub has generally available the redesigned Copilot CLI terminal interface, introducing a tabbed layout for managing sessions, gists, issues, and pull requests. The update features a form-driven setup process for MCP servers, skills, and plugins, eliminating the need to manually edit configuration files. Additionally, the UI has been refined for better accessibility, including screen reader support and theme awareness.

• Tabbed layout consolidates sessions, gists, issues, and PRs in a single terminal view.
• Form-driven setup for MCP servers and plugins removes manual config file editing.
• Improved accessibility includes screen reader support and theme-aware design.
• Terminal interface is now generally available for all users.