Feeds

In this post, we explore how Instagram has successfully scaled its algorithm to include over 1000 ML models without sacrificing recommendation quality or reliability. We delve into the intricacies of managing such a vast array of models, each with its own performance characteristics and product goals. We share insights and lessons learned along the way—from the initial realization that our infrastructure maturity was lagging behind our ambitious scaling goals, to the innovative solutions we implemented to bridge these gaps. In the ever-evolving landscape of social media, Instagram serves as a hub for creative expression and connection, continually adapting to meet the dynamic needs of its global community. At the heart of this adaptability lies a web of machine learning (ML) models, each playing a crucial role in personalizing experiences. As Instagram’s reach and influence has grown, so too has the complexity of its algorithmic infrastructure. This growth, while exciting, presents a unique set of challenges, particularly in terms of reliability and scalability. ...

I’m in Silicon Valley today to get ready for Google I/O which starts tomorrow. It’s a welcome distraction as I had a complex weekend. Some great family time, but then one of my two dogs (Ellie) passed away. I loved that pup and had almost thirteen great years of companionship. Dogs are the best. [blog] Vibe Coding a 5,000km Bike Race: Building a Route Planner with Firebase & Gemini. Cool post from a colleague who had a use case, and quickly solved it with AI assistance. ...

Top Links VS Code: Open Source AI Editor (VS Code Team) Introducing Windows ML: The future of machine learning development on Windows (Tucker Burns) ReSharper Comes to Microsoft Visual Studio Code: Public Preview Now Open (Sasha Ivanova) Windows App SDK v1.8.0 Experimental 2 Release (1.8.250515001-experimental2) (Roy MacLachlan) Advancing Windows for AI development: New platform capabilities and tools introduced at Build 2025 (Pavan Davuluri) Agent mode for every developer (Katie Savage) The Windows Subsystem for Linux is now open source (Pierre Boulay) Enhance your application security with administrator protection (Nilanjana Ganguly) Accelerating AI development with Windows-based AI workstations (Shen Chauhan) Accelerate Your .NET Upgrades with GitHub Copilot (McKenna Barlow) Announcing SQL Server 2025 Public Preview (Bob Ward) Post-Quantum Cryptography Comes to Windows Insiders and Linux (Aabha Thipsay) Securing the Model Context Protocol: Building a safer agentic future on Windows (David Weston) Edit is now open source (Christopher Nguyen) Microsoft Store expands opportunities for Windows app developers (Giorgio Sardo) .NET Aspire 9.3 is here and enhanced with GitHub Copilot! (Jeffrey T. Fritz) ...

😘 Kiss bugs goodbye with fully automated end-to-end test coverage (Sponsored)Bugs sneak out when less than 80% of user flows are tested before shipping. However, getting that kind of coverage (and staying there) is hard and pricey for any team. QA Wolf’s AI-native service provides high-volume, high-speed test coverage for web and mobile apps, reducing your organizations QA cycle to less than 15 minutes. ...

The HashiCorp Technical Field Organization (TFO) partners with enterprise customers to help them navigate complex infrastructure challenges and achieve specific technical outcomes. Over thousands of these engagements, consistent patterns have emerged. We’ve documented many of these battle-tested patterns using HashiCorp products and the broader tooling ecosystem to solve common challenges. We call these HashiCorp Validated Patterns and we’re now making them publicly available. One of the unique values of Validated Patterns is that they reflect shared knowledge across our global technical field staff. These solutions aren’t the opinion of one person. They’re reviewed, improved, and endorsed by our internal technical community. ...

As Meta has launched new, innovative products leveraging generative AI (GenAI), we need to make sure the underlying infrastructure components evolve along with it. Applying infrastructure knowledge and optimizations have allowed us to adapt to changing product requirements, delivering a better product along the way. Ultimately, our infrastructure systems need to balance our need to ship high-quality experiences with a need to run systems sustainability. Splitting GenAI inference traffic out into a dedicated WWW tenant, which allows specialized runtime and warm-up configuration, has enabled us to meet both of those goals while delivering a 30% improvement in latency. ...

⚠️ WARNING ⚠️ This blog post contains graphic depictions of probability. Reader discretion is advised. Measuring performance is tricky. You have to think about accuracy and precision. Are your sampling rates high enough? Could they be too high?? How much metadata does each recording need??? Even after all that, all you have is raw data. Eventually for all this raw performance information to be useful, it has to be aggregated and communicated. Whether it's in the form of a dashboard, customer report, or a paged alert, performance measurements are only useful if someone can see and understand them. ...

In this post I show how you can push a whole stack of branches with a single command. This is particularly useful when you’re working with stacked branches, and want to push all the branches in the stack at once. In this post I show the git aliases I have created that make doing this a single, simple command. What are stacked branches? I’m a big fan of creating small git commits. Especially when you’re creating a big feature. I like to create a “story” with my commits, adding bits of a larger feature commit by commit. The idea is to make it as simple as possible for others to review by walking through a commit at a time. ...