InfoQ

How Observability and Telemetry Can Enhance the Practice of Software Engineering

Ben Linders — Thu, 23 Apr 2026 11:04:00 GMT

Observability must evolve with serverless, event-driven architectures. OpenTelemetry can decouple telemetry from vendors, letting developers emit consistent, high-quality data that explains real system behavior. Shared vocabularies and good telemetry make debugging faster and improve reliability, speed, and developer productivity.

By Ben Linders

Presentation: How to Build an Exchange: Sub Millisecond Response Times and 24/7 Uptimes in the Cloud

Frank Yu — Thu, 23 Apr 2026 09:29:00 GMT

Frank Yu shares Coinbase’s engineering philosophy for building resilient, fair, and fast financial exchanges. He explains the power of a single-threaded architecture combined with the Raft consensus algorithm to maintain 24/7 availability. He discusses how determinism enables zero-downtime rolling deployments and the ability to replay production logs for perfect bug reproduction.

By Frank Yu

Dropbox Collaborates with GitHub to Reduce Monorepo Size from 87GB to 20GB

Leela Kumili — Wed, 22 Apr 2026 14:14:00 GMT

Dropbox reduced its backend monorepo from 87GB to 20GB by optimizing Git delta compression in collaboration with GitHub. The changes improved clone times, CI performance, and developer velocity, highlighting how repository storage inefficiencies can impact large-scale engineering workflows.

By Leela Kumili

Presentation: Panel: Building a Culture that Works

Wed, 22 Apr 2026 10:10:00 GMT

The panelists share insights on evolving company culture. They discuss leveraging feedback loops, lending social capital, and the friction between legacy bureaucracy and agile engineering. The panel explains how to maintain cohesion in remote teams and use interviews to uncover the true "unmanicured" culture of a firm.

By Nicky Wrightson, Suhail Patel, Lesley Cordero, Matthew Card, Natan Žabkar Nordberg

Cloudflare Sandboxes Reach General Availability, Giving AI Agents Persistent Isolated Environments

Steef-Jan Wiggers — Wed, 22 Apr 2026 10:00:00 GMT

Cloudflare has released Sandboxes and Containers into general availability, providing persistent isolated Linux environments for AI agent workloads. New capabilities include secure credential injection via egress proxy, PTY terminal support, persistent code interpreters, filesystem watching, and snapshot-based session recovery. Active CPU pricing charges only for used cycles.

By Steef-Jan Wiggers

Article: When a Cloud Region Fails: Rethinking High Availability in a Geopolitically Unstable World

Rohan Vardhan — Wed, 22 Apr 2026 09:00:00 GMT

Sovereign fault domains are failure boundaries defined by legal, political, or physical jurisdiction rather than hardware topology. The article maps geopolitical events to known distributed-systems failure modes, argues multi-region should replace multi-AZ as the HA baseline for systems crossing jurisdictions, and outlines design patterns, chaos experiments, and an ALE model to justify the spend.

By Rohan Vardhan

Cloudflare Outlines MCP Architecture as Enterprises Confront Security and Governance Risks

Matt Foster — Wed, 22 Apr 2026 07:38:00 GMT

Cloudflare has outlined a reference architecture for scaling Model Context Protocol (MCP) deployments across the enterprise, positioning centralized governance, remote server infrastructure, and cost controls as key requirements for production-ready agent systems.

By Matt Foster

pnpm 11 Release Candidate: ESM Distribution, Supply Chain Defaults and a New Store Format

Daniel Curtis — Tue, 21 Apr 2026 16:00:00 GMT

pnpm 11 RC has been released, featuring significant changes in performance, security, and configuration. Key updates include an SQLite-backed store index, tighter security defaults, and a consolidated build script setting. It now requires Node.js v22 or later. Global installs are isolated by default, and new commands enhance usability. Migration guidance is available in the documentation.

By Daniel Curtis

Anthropic Introduces Managed Agents to Simplify AI Agent Deployment

Leela Kumili — Tue, 21 Apr 2026 14:36:00 GMT

Anthropic introduces Managed Agents on Claude, a managed execution layer for agent-based workflows. It separates agent logic from runtime concerns like orchestration, sandboxing, state management, and credentials. The system supports long-running multi-step workflows with external tools, error recovery, and session continuity via a meta-harness architecture.

By Leela Kumili

Slack Rebuilds Notification System, Reports 5X Increase in Settings Engagement

Leela Kumili — Tue, 21 Apr 2026 14:05:00 GMT

Slack has rebuilt its notification system with a unified architecture that separates activity from delivery, improving consistency across platforms. The redesign simplifies preferences, preserves legacy settings through transformation, and resulted in a 5x increase in user engagement with notification settings along with reduced support tickets.

By Leela Kumili

GitHub Acknowledges Recent Outages, Cites Scaling Challenges and Architectural Weaknesses

Craig Risi — Tue, 21 Apr 2026 12:00:00 GMT

GitHub has publicly addressed a series of recent availability and performance issues that disrupted services across its platform, attributing the incidents to rapid growth, architectural coupling, and limitations in handling system load.

By Craig Risi

Presentation: Dynamic Moments: Weaving LLMs into Deep Personalization at DoorDash

Sudeep Das, Pradeep Muthukrishnan — Tue, 21 Apr 2026 10:35:00 GMT

Sudeep Das and Pradeep Muthukrishnan explain the shift from static merchandising to dynamic, moment-aware personalization at DoorDash. They share how LLMs generate natural-language "consumer profiles" and content blueprints, while traditional deep learning handles last-mile ranking. This hybrid approach allows the platform to adapt to short-lived user intent and massive catalog abundance.

By Sudeep Das, Pradeep Muthukrishnan

Article: Redesigning Banking PDF Table Extraction: A Layered Approach with Java

Mehuli Mukherjee — Tue, 21 Apr 2026 09:00:00 GMT

PDF table extraction often looks easy until it fails in production. Real bank statements can be messy, with scanned pages, shifting layouts, merged cells, and wrapped rows that break standard Java parsers. This article shares how we redesigned the approach using stream parsing, lattice/OCR, validation, scoring, and selective ML to make extraction more reliable in real banking systems.

By Mehuli Mukherjee

Cloudflare Introduces Project Think: a Durable Runtime for AI Agents

Patrick Farry — Tue, 21 Apr 2026 02:34:00 GMT

Cloudflare's Project Think introduces a new framework for AI agents, shifting from stateless orchestration to a durable actor-based infrastructure. It features a kernel-like runtime enabling agents to manage memory and run code securely. Innovations include Fibers for checkpointing progress and a Session API for relational conversations, enhancing agent efficiency and resilience.

By Patrick Farry

Designing Memory for AI Agents: inside Linkedin’s Cognitive Memory Agent

Leela Kumili — Mon, 20 Apr 2026 14:59:00 GMT

LinkedIn introduces Cognitive Memory Agent (CMA), generative AI infrastructure layer enabling stateful, context-aware systems. It provides persistent memory across episodic, semantic, and procedural layers, supporting multi-agent coordination, retrieval, and lifecycle management. CMA addresses LLM statelessness and enables production-grade personalization and long-term context in AI applications.

By Leela Kumili