Dmitri Riz

The Power Platform custom connector I created for the U.S. Congress CRS is now available in all regions. This means that Power Apps, Power Automate, and Logic Apps users can tap directly into the authoritative Congressional Research Service (CRS) dataset. Works with Copilot as well. Link to the connector docs in the comments. (Not available in GCC or GCC High environments, womp womp 🤷) #PowerPlatform #CustomConnector #CRS #CitizenDevelopment #GovTech #Automation

22

1 Comment

New PeopleSoft idea you may want to upvote: Put an App Class "Builder" button next to the Loader Class field of Landing Page sections to help developers avoid typos and select the correct App Class (similar to Event Mapping, Tile Wizard, etc.). https://lnkd.in/gwfBUGV4

31

This section of the article really stuck out to me: "The generated code often required substantial manual intervention. Performance problems emerged. Maintenance became a nightmare when generated code diverged from the visual models. Most critically, drawing accurate diagrams required understanding the same logical complexity that programming demanded. The tool changed the interface but not the fundamental challenge." That excerpt is talking about CASE Tools from the 1980s. But doesn't that sound like commentary that could be made about today's AI innovations? This is an example of why I focus on helping developers be more efficient. Developers need more time to reason through the logical complexity that arises from human beings and the many permutations of activity they create.

8

1 Comment

Part 2 is live: C.R.A.F.T. Framework with .NET and Azure I broke down exactly how we're using systematic AI-assisted development on my current restaurant tech platform. Key insights: ✅ Multi-agent orchestration: Claude Code for complex logic, Copilot for patterns, GPT-5 for troubleshooting ✅ Centralized documentation that all AI agents reference ✅ GitHub Copilot's Azure integration providing real-time infrastructure context ✅ Multi-layer security scanning catching issues automatically ✅ What actually breaks (file sprawl, multi-tenant logic gaps, performance assumptions) The reality: Different AI models excel at different tasks. The teams getting 3-4x velocity aren't using one tool—they're orchestrating multiple agents with consistent documentation and validation. This is production implementation, not proof of concept. Multi-repository architecture. Real security requirements. Actual scaling constraints. Article: https://buff.ly/MPRBRvL What's your experience with multi-agent AI development? #SoftwareDevelopment #ArtificialIntelligence #DotNet #Azure #DevSecOps

2

Anthropic dropped its first official technical certification last week and I think the exam blueprint tells you more than the announcement did. The Claude Certified Architect (CCA) Foundations launched March 12. It's a 60-question proctored exam. No Claude, no docs, no external tools during the test. Passing score is 720 on a 1000-point scale. You get 120 minutes. Here's what's actually on it: Agentic Architecture & Orchestration — 27% Claude Code Configuration & Workflows — 20% Prompt Engineering & Structured Output — 20% Tool Design & MCP Integration — 18% Context Management & Reliability — 15% That 27% on agentic architecture is the tell. This isn't about knowing API parameters. The questions drop you into production scenarios — a customer support agent hitting 55% first-contact resolution when the target is 80%, a 14-file PR review producing inconsistent results, a batch vs. real-time API decision under latency constraints. You make the architectural call. The exam pulls from 6 scenarios (customer support agent, multi-agent research system, Claude Code for CI/CD, structured data extraction and two more). Only 4 are shown to any candidate. Study all 6 because each one overlaps multiple domains. Right now it's gated behind the Claude Partner Network, which any organization can join free. The first 5,000 partner-company employees get early access at no cost. After that it's $99 a sitting. The 13 prep courses on Anthropic Academy via Skilljar are free and open to anyone regardless of partner status. More certifications are coming later this year for sellers, developers and advanced architects. This is the entry point of a stack, not a standalone badge. The format matters. A proctored exam with a scaled score carries more weight than a Skilljar completion badge. Accenture has already embedded this as a baseline expectation for Claude delivery roles across 30,000 professionals. If you're building production Claude integrations and not just prototyping, this one is worth your time. https://lnkd.in/egX7Vdv3 #claudeCertifiedArchitect #AI #MCP #Anthropic #PromptEngineering #Exam

6

1 Comment

We have discovered that an issue can occur with #sqlserver2025 if you have a large number of databases for your instance. We have documented this problem at https://lnkd.in/g-SmMk3r. On this page you will see a workaround to use a startup trace flag to avoid the problem. You are fully supported in production to use this until we produce a fix in a future cumulative update. Feel free to comment if you have questions about this. #sqlserver

319

21 Comments

Excellent Tutorial on Parsing JSON Data in Power Apps: A Step-by-Step Tutorial. One thing that gave me a major problem: Under Power App Settings =>Support => Edit "Authoring version" Ensure you have the Recommended Version or the latest. My ParseJSON function would not parse JSON until I changed the Authoring Version. https://lnkd.in/efxKgPzQ #ParseJSON #ConvertingJSON

1

Looking to move to IBM Git/DBB from Endevor, Changeman, or SCLM? Consider a two stage compile process, and be sure to review your compile parameters to optimize for both debugging and production load performance! https://lnkd.in/eNAH3qHa #mainframe #cobol #zos

35

📢 The NEW Grid Container in Power Apps introduces a new way to build responsive Canvas App layouts. 🎥 Watch here: https://lnkd.in/gRZHE3h2 In this video, I explore the GRID Container control with examples and more. #PowerApps #CanvasApps #ResponsiveDesign

695

19 Comments

This note from Graham Cunningham on Anthropic announcement on COBOL language conversion is to the point. This observation comes from CTO of Heirloom Computing whose main business is to migrate from mainframe and who uses Claude extensively in their product. "Engineering at scale has one non-negotiable requirement: given the same input, you must get the same output. 𝐄𝐯𝐞𝐫𝐲 𝐭𝐢𝐦𝐞. 𝐖𝐢𝐭𝐡𝐨𝐮𝐭 𝐞𝐱𝐜𝐞𝐩𝐭𝐢𝐨𝐧."

10

2 Comments

⚡ .NET 10 is 21% Faster? My .NET 8/9/10 Benchmark Results I initially started benchmarking .NET 9 just to get a feel for it in my daily work 🧑💻. But with the release of .NET 10, I couldn’t resist doing a proper deep dive 🔍. So I built a comprehensive, side-by-side comparison of .NET 8 vs 9 vs 10 on the same machine to see what the real-world numbers actually look like ⚙️. The results were honestly surprising 😮. While I expected improvements, seeing .NET 10 clock in at ~21% faster overall compared to .NET 8 was impressive 🔥. A few key takeaways that stood out to me: - JSON is blazing fast 🚀: .NET 10 delivers nearly 25% faster JSON operations a huge win for any API-heavy architecture. - Memory efficiency 🧠: Memory allocation is roughly 44% faster in .NET 10 compared to .NET 8. - It’s not a clean sweep 🤝: .NET 9 still holds the crown for LINQ and some collection-heavy scenarios, but .NET 10 takes the lead for overall execution speed and “tight-loop” style workloads. I originally did this research to support our own engineering decisions at work, but the data felt too useful to keep private — so I decided to share it with everyone ❤️. You can grab the code to run it yourself or read the full report here: 👉 Repository - (https://lnkd.in/deU3Hgrt) 📊 Full Performance Report - (https://lnkd.in/dGWSSqyv) Hope this helps anyone planning their next migration ✨ --- #dotnet #csharp #performance #benchmarking #systemtextjson #jit

1