An anonymous reader shares a report: The most seductive narrative in American work culture right now isn't that AI will take your job. It's that AI will save you from it. That's the version the industry has spent the last three years selling to millions of nervous people who are eager to buy it. Yes, some white-collar jobs will disappear. But for most other roles, the argument goes, AI is a force multiplier. You become a more capable, more indispensable lawyer, consultant, writer, coder, financial analyst -- and so on. The tools work for you, you work less hard, everybody wins.
But a new study published in Harvard Business Review follows that premise to its actual conclusion, and what it finds there isn't a productivity revolution. It finds companies are at risk of becoming burnout machines.
As part of what they describe as "in-progress research," UC Berkeley researchers spent eight months inside a 200-person tech company watching what happened when workers genuinely embraced AI. What they found across more than 40 "in-depth" interviews was that nobody was pressured at this company. Nobody was told to hit new targets. People just started doing more because the tools made more feel doable. But because they could do these things, work began bleeding into lunch breaks and late evenings. The employees' to-do lists expanded to fill every hour that AI freed up, and then kept going.
The SQL community is gathering in Atlanta this March for the first‑ever SQLCon, co‑located with FabCon, the Microsoft Fabric Community Conference, March 16-20. One registration unlocks both events, giving you access to deep SQL expertise and the latest in Fabric, Power BI, data engineering, real‑time intelligence, and AI. Whether you’re a DBA, developer, data engineer, architect, or a leader building data‑driven team, this is your chance to learn, connect, and shape what’s next.
One pass, two conferences—double the value Register once, benefit twice. With SQLCon and FabCon under the same roof, you can mix a deep SQL session in the morning with a Fabric or AI talk in the afternoon, then drop into the shared expo and community lounge. It’s a seamless, high‑impact week that lets specialists go deep while cross‑functional teams build a common language across data, analytics, and AI.
Dive Deep with Interactive Sessions and Hands-On Workshops There are 50 SQL sessions at SQLCon. Fifty! The program is designed for momentum. Across the week, you’ll find practical content on SQL Server, Azure SQL, SQL database in Fabric, performance tuning, security and governance, migration and modernization, and building AI‑powered experiences with SQL. Monday and Tuesday are hands‑on workshop days—bring your laptop and leave with repeatable scripts, patterns, and demos you can apply immediately. Wednesday through Friday, you’ll stack conference sessions to round out your plan for the year.
Experience Atlanta: The Perfect Setting for SQLCon SQLCon + FabCon take place at the Georgia World Congress Center, in the heart of a walkable downtown that’s tailor‑made for a great conference week. You’ll be just a short walk from Centennial Olympic Park, near State Farm Arena—home to major keynote events—and amid lively dining and music options. The attendee party is at the Georgia Aquarium, an unforgettable after‑hours experience with spectacular exhibits and a perfect setting for relaxed conversations with peers and product teams. Want a quick vibe check on the city and the conference energy? Watch the short video of Guy in a Cube and me:
Announcements on roadmap, engineering insights, and live updates If you want to understand where SQL Server, Azure SQL, and SQL database in Fabric are heading, this is the place. Expect direct updates from engineering (we’re sending over 30 members from the SQL product team); first‑look announcements; and live demos of upcoming capabilities across SQL tooling and drivers, SSMS/VS Code extensions, Copilot integrations, and Fabric SQL experiences. You’ll leave with clarity on what’s coming, how it impacts your environment, and where to invest next.
The SQL Community: Revitalized and Engaged SQLCon goes beyond a conference—it’s a gathering where the lounge hosts meetups and active conversations. Ask‑the‑Experts sessions connect you with engineers, MVPs, and product teams. Shared keynotes bring everyone together, and the city makes it easy to extend conversations into the evening. Bring your toughest questions, real-world challenges, and bold goals—you’ll leave with practical solutions, valuable connections, and new inspiration.
Bonus: make the budget work Depending on timing, look for early‑bird pricing, team discounts, or buy‑one‑get‑one offers on the registration page. These deals move fast, so check what’s live when you register. You can always use SQLCMTY200 for $200 off!
Wrap‑up: build the next chapter of your data strategy at SQLCon SQLCon + FabCon is the highest‑leverage week of the year to sharpen your technical skills, understand SQL’s next chapter, accelerate modernization and performance, and build meaningful connections across the global community. If SQL plays any role in your data estate, this is the one event you shouldn’t miss.
We're inviting all developers to join Agents League, running February 16-27. It's a two-week challenge where you'll build AI agents using production-ready tools, learn from live coding sessions, and get feedback directly from Microsoft product teams.
We've put together starter kits for each track to help you get up and running quickly that also includes requirements and guidelines. Whether you want to explore what GitHub Copilot can do beyond autocomplete, build reasoning agents on Microsoft Foundry, or create enterprise integrations for Microsoft 365 Copilot, we have a track for you.
Important:Register first to be eligible for prizes and your digital badge. Without registration, you won't qualify for awards or receive a badge when you submit.
What Is Agents League?
It's a 2-week competition that combines learning with building:
📽️ Live coding battles – Watch Product teams, MVPs and community members tackle challenges in real-time on Microsoft Reactor
💻 Async challenges – Build at your own pace, on your schedule
💬 Discord community – Connect with other participants, join AMAs, and get help when you need it
🏆 Prizes – $500 per track winner, plus GitHub Copilot Pro subscriptions for top picks
The Three Tracks
🎨 Creative Apps — Build with GitHub Copilot (Chat, CLI, or SDK)
More details on each track below, or jump straight to the starter kits.
The Schedule
Agents League starts on February 16th and runs through Feburary 27th. Within 2 weeks, we host live battles on Reactor and AMA sessions on Discord.
Week 1: Live Battles (Feb 17-19)
We're kicking off with live coding battles streamed on Microsoft Reactor. Watch experienced developers compete in real-time, explaining their approach and architectural decisions as they go.
Tue Feb 17, 9 AM PT — 🎨 Creative Apps battle
Wed Feb 18, 9 AM PT — 🧠 Reasoning Agents battle
Thu Feb 19, 9 AM PT — 💼 Enterprise Agents battle
All sessions are recorded, so you can watch on your own schedule.
Week 2: Build + AMAs (Feb 24-26)
This is your time to build and ask questions on Discord. The async format means you work when it suits you, evenings, weekends, whatever fits your schedule.
We're also hosting AMAs on Discord where you can ask questions directly to Microsoft experts and product teams:
Tue Feb 24, 9 AM PT — 🎨 Creative Apps AMA
Wed Feb 25, 9 AM PT — 🧠 Reasoning Agents AMA
Thu Feb 26, 9 AM PT — 💼 Enterprise Agents AMA
Bring your questions, get help when you're stuck, and share what you're building with the community.
Pick Your Track
We've created a starter kit for each track with setup guides, project ideas, and example scenarios to help you get started quickly.
🎨 Creative Apps
Tool: GitHub Copilot (Chat, CLI, or SDK)
Build innovative, imaginative applications that showcase the potential of AI-assisted development. All application types are welcome, web apps, CLI tools, games, mobile apps, desktop applications, and more.
The starter kit walks you through GitHub Copilot's different modes and provides prompting tips to get the best results. View the Creative Apps starter kit.
🧠 Reasoning Agents
Tool: Microsoft Foundry (UI or SDK) and/or Microsoft Agent Framework
Build a multi-agent system that leverages advanced reasoning capabilities to solve complex problems. This track focuses on agents that can plan, reason through multi-step problems, and collaborate.
The starter kit includes architecture patterns, reasoning strategies (planner-executor, critic/verifier, self-reflection), and integration guides for tools and MCP servers. View the Reasoning Agents starter kit.
💼 Enterprise Agents
Tool: M365 Agents Toolkit or Copilot Studio
Create intelligent agents that extend Microsoft 365 Copilot to address real-world enterprise scenarios. Your agent must work on Microsoft 365 Copilot Chat.
To be eligible for prizes and your digital badge, you must register before submitting your project.
Category Winners ($500 each):
🎨 Creative Apps winner
🧠 Reasoning Agents winner
💼 Enterprise Agents winner
GitHub Copilot Pro subscriptions:
Community Favorite (voted by participants on Discord)
Product Team Picks (selected by Microsoft product teams)
Everyone who registers and submits a project wins: A digital badge to showcase their participation.
Beyond the prizes, every participant gets feedback from the teams who built these tools, a valuable opportunity to learn and improve your approach to AI agent development.
How to Get Started
Register first — This is required to be eligible for prizes and to receive your digital badge. Without registration, your submission won't qualify for awards or a badge.
Pick a track — Choose one track. Explore the starter kits to help you decide.
Watch the battles — See how experienced developers approach these challenges. Great for learning even if you're still deciding whether to compete.
Build your project — You have until Feb 27. Work on your own schedule.
Agentic AI is best understood as a distributed system, and many of the same patterns that made microservices successful apply. This article explains how to design agentic workflows with composable components, explicit contracts and guardrails, resilience practices like timeouts and idempotency, and end-to-end observability. We will also discuss how the Red Hat AI portfolio supports production-ready agentic systems across the hybrid cloud, from efficient inference to consistent execution and lifecycle management at scale.
Agentic AI through a microservices lens
Microservices changed software engineering by forcing us to treat systems as distributed by default with small components, explicit contracts, independent scaling, and a serious focus on reliability and observability. Agentic AI is driving a similar shift. Instead of HTTP services calling other services, we now have agents coordinating models, tools, and enterprise data to complete multi-step tasks.
If you build cloud-native applications, this analogy is useful because it replaces hand-wavy intuition with a familiar engineering frame. Agentic AI is a distributed system. The same realities apply: latency, partial failure, versioning, security boundaries, and operational visibility. Once you accept that, building agentic systems becomes far more practical. Figure 1 illustrates a comparison of microservices patterns and Agentic AI equivalents.
Figure 1: This chart compares microservices patterns and Agentic AI equivalents, aligning common concepts such as contracts, decomposition, resiliency, observability, scaling, and governance.
From one big agent to composable workflows
Early agent implementations often become a single, do-everything agent. It retrieves context, decides what to do, calls tools, handles errors, and writes the final answer. That looks convenient, but it is the AI equivalent of a monolith. When something goes wrong, it is difficult to isolate the cause. When you want to improve one part (i.e., retrieval), you risk breaking everything.
A more scalable pattern is the same one microservices pushed us toward, decomposition. Break the workflow into smaller, purpose-built agents and orchestrate them as a pipeline. For example, you might have an agent that retrieves and ranks information, another that validates policy and safety constraints, and another that executes tool calls and formats results. You can test, update, and scale each component independently, so that failures become easier to contain.
Contracts are more important with agents
Microservices succeed when interfaces are explicit. Agents need that discipline even more, because ambiguity is where unpredictable behavior lives. Define what the agent accepts and produces, ideally with structured outputs you can validate (e.g., JSON schema). You must be equally explicit about tool contracts and allowlists. For instance, define what tools can be called, with what parameters, and what data can be accessed. This is how you prevent prompt drift from turning into system drift, and it is how you make agentic workflows governable across teams.
A strong contract mindset also improves portability. When your tools and agent steps have stable interfaces, you can swap models, change retrieval methods, or add new workflow steps without rewriting the whole system.
Reliability patterns carry over
Microservices taught us to assume failure. Networks drop, dependencies degrade, and tail latency ruins user experience. Agentic systems have the same issues plus a few new ones, such as tool calls failing, retrieval returning irrelevant context, and inference latency spikes. The lesson from microservices still holds. It is the same operational playbook: timeouts, retries with backoff, circuit breakers to avoid cascading failures, and fallback behaviors that let a workflow degrade gracefully. For tool calls, it also helps to design for idempotency so retries do not create duplicate actions.
This is where agentic design becomes an engineering discipline. You define failure semantics, set performance expectations, and decide what the system should do under degradation. In other words, you are making explicit tradeoffs about latency, cost, and correctness when a dependency is slow or unavailable.
Observability: Tracing decisions
In microservices, we trace requests through services. In agentic systems, we also need to trace decisions through workflow steps. When a result is wrong, you want to know whether the failure came from retrieval, a tool invocation, or the model’s reasoning. That means capturing step-level traces, tool-call inputs and outputs, and inference performance metrics; then correlating them end-to-end through a single trace that spans the workflow.
Without observability, agentic AI remains stuck in the prototype stage. With it, teams can tune quality, reduce cost, and improve reliability with the same confidence they bring to cloud-native operations.
Figure 2 shows a simple reference architecture for an agentic workflow, tracing a request from the client through an orchestrator and agent services, into tools and data, then model inference, and finally end-to-end observability.
Figure 2: Reference architecture for agentic AI workflows.
How Red Hat AI fits in this microservices-inspired model
Agentic AI becomes real when it is supported by a consistent platform layer across hybrid cloud, one that delivers efficient inference, consistent runtimes, and lifecycle management at scale. Together, the following Red Hat services map cleanly to the microservices mindset, including inference as a service, consistent runtimes where you need them, and a platform to build and operate AI workflows reliably across environments.
Red Hat AI Inference Server provides the inference layer by turning model execution into a managed, high-throughput service. Powered by vLLM, it helps maximize accelerator utilization and reduce latency so agentic workflows can make frequent model calls without cost or performance surprises. It also provides access to validated, optimized third-party models, helping teams standardize their deployments across environments.
Red Hat Enterprise Linux AI provides a purpose-built, single-server platform for AI inference workloads, such as LLMs, when tighter control and repeatable operations matter. It combines a bootable Red Hat Enterprise Linux image with popular AI libraries and hardware-optimized inference, and it includes Red Hat AI Inference Server so teams can start serving models quickly with a consistent, supported stack.
Red Hat OpenShift AI provides an operationally consistent platform to build, train, fine-tune, serve, and monitor predictive and generative AI models at scale across hybrid cloud environments, including private and sovereign deployments. It includes capabilities for model monitoring and drift detection, and it supports distributed serving through an optimized vLLM framework while standardizing access to models and tools for agentic workflows.
This mindset shift unlocks production
Agentic AI is not a replacement for engineering discipline. It demands more of it. Agentic AI is a distributed system with probabilistic components, and that raises the bar for architecture and operations. In practice, you must engineer agents like microservices by breaking workflows into composable components, enforcing clear contracts and guardrails, designing for failure with timeouts and fallbacks, and instrumenting everything so behavior is observable and auditable.
Using IndexOf() for checking if a string starts with a specific value in .NET is inefficient and unclear. This approach performs unnecessary work and obscures intent.
Secure password hashing is vital for application security. The PasswordHasher class in Spargine for .NET 10 consolidates previous hashing implementations into a single solution that supports various modern algorithms. This enhances flexibility and ensures robust security while simplifying usage, making it essential for developers aiming to protect user credentials effectively.
Log in
