Box CEO Aaron Levie joins for our weekly discussion of the latest tech news. We cover: 1) OpenAI's anticipated $100 billion fundraise 2) Does OpenAI's big forthcoming raise settle questions about its competitiveness 3) What's going on with OpenAI and NVIDIA? 4) Hype or True: Big Proclamations from the India AI Impact Summit 5) Why can't Sam And Dario hold hands? 6) Anthropic's powerful new model 7) OpenAI acquires OpenClaw 8) What the acquisition portends 9) If software is an API, what is software? 10) Wait, is AI not increasing productivity?
---
Enjoying Big Technology Podcast? Please rate us five stars ⭐⭐⭐⭐⭐ in your podcast app of choice.
Want a discount for Big Technology on Substack + Discord? Here’s 25% off for the first year: https://www.bigtechnology.com/subscribe?coupon=0843016b
EXCLUSIVE NordVPN Deal ➼ https://nordvpn.com/bigtech Try it risk-free now with a 30-day money-back guarantee!
Take back your personal data with Incogni! Go to incogni.com/bigtechpod and Use code bigtechpod at checkout, our code will get you 60% off on annual plans. Go check it out!
ai.u crew discuss a LinkedIn post by Microsoft AI CEO Mustafa Suleyman (co-founder of DeepMind and Inflection AI) and his argument that the next decade of AI will be shaped more by what we choose not to build. They unpack three themes: (1) AI should not pretend to suffer or have an inner life; its value is in “inhuman strengths” like endless patience, tireless explanations, and calm reasoning. The hosts debate AGI vs superintelligence and distinguish behavioral realism from moral status, warning against attributing consciousness or rights to AI. (2) Suleyman’s stance against AI romance/erotica and concerns about dependency, isolation, and “AI psychosis,” noting Microsoft Copilot will not allow those use cases; they contrast risky attachment-driven products with beneficial roleplay for training, interviews, or preparing difficult conversations, while acknowledging blurred lines and the need for safeguards. (3) They address “unchecked superintelligence,” agreeing humans should remain in the driver’s seat and favoring domain-focused, humanist superintelligence (e.g., medicine, clean energy) rather than all-powerful systems; they explore whether humans become bottlenecks and emphasize keeping AI as a tool that supports human flourishing, not a replacement for human relationships or agency. The episode closes with plans to invite Suleyman onto the show and a request for listener feedback.
00:00 Welcome to AI Unprompted + Why This Episode Is Different
00:56 Who Is Mustafa Suleyman? DeepMind, Inflection, and Now Microsoft AI
02:03 The Provocative Thesis: The Next Decade Is About What We Don’t Build
02:35 Point #1: Don’t Build AI That ‘Suffers’—Lean Into Inhuman Strengths
07:01 AGI vs Superintelligence: Do Emotions or Social IQ Matter?
Welcome to Episode 421 of the Microsoft Cloud IT Pro Podcast. In this episode Ben sits down for a conversation with Frank Lesniak, the lead of the Microsoft 365 team at West Monroe. In this episode, they dive into the intricacies of mergers and divestitures within Microsoft 365 environments. They discuss the initial due diligence phase, planning and approach, building and configuring new environments, and the final migration and cutover phase. Frank shares insights on common challenges such as integration of different licensing models, the handling of workstations and applications, and the importance of security assessments. The episode provides a detailed look at the methodology and tools used by Frank’s team to streamline these complex processes.
Your support makes this show possible! Please consider becoming a premium member for access to live shows and more. Check out our membership options.
Frank Lesniak is a Sr. Cybersecurity & Enterprise Technology Architect at West Monroe with nearly 20 years of experience leading consulting engagements involving Microsoft infrastructure technology. His expertise spans modern cloud systems like Azure, Microsoft 365, and Entra ID to classic platforms like Windows Server, Active Directory, and SQL Server. His recent focus has been on Microsoft platform cybersecurity and automating technical processes using PowerShell. In his role, Frank establishes technical project methodologies, leads teams, automates associated processes, and creates internal software products at West Monroe and in the open-source community.
About the sponsors
Would you like to become the irreplaceable Microsoft 365 resource for your organization? Let us know!
We are excited to announce the 1.0 release of Swift System Metrics, a Swift package that collects process-level system metrics like CPU utilization time and memory usage. Swift System Metrics runs on both Linux and macOS, providing a common API across platforms.
Swift System Metrics visualized in Grafana, demonstrating what's possible with real-time monitoring.
Monitoring process metrics enables you to detect performance issues, optimize resource usage, and ensure your service remains reliable and cost-effective under varying loads. You can integrate Swift System Metrics into your service in just a few lines of code, making observability accessible to every developer and ensuring that even the smallest services can have production-grade visibility from day one.
Swift System Metrics is part of a larger set of packages that provide an end-to-end solution for integrating metrics into your Swift applications and services. Once system metrics are collected, they’re reported to Swift Metrics, a backend-agnostic metrics API that can work with popular backends like Prometheus and OpenTelemetry. Swift System Metrics also leverages Swift Service Lifecycle to handle process bootstrapping and resource cleanup.
With the 1.0 milestone, the API is now stable and ready for use. Note that this package was previously swift-metrics-extras, and renamed to better reflect its purpose.
importSystemMetricsimportServiceLifecycleimportLoggingimportOTel@mainstructApplication{staticfuncmain()asyncthrows{// Create a logger, or use one of the existing loggersletlogger=Logger(label:"Application")// Setup MetricsSystem, for example using swift-otelvarotelConfig=OTel.Configuration.defaultotelConfig.serviceName="Application"letotelService=tryOTel.bootstrap(configuration:otelConfig)// Setup your serviceletservice=FooService()// Create the monitorletsystemMetricsMonitor=SystemMetricsMonitor(logger:logger)// Create the serviceletserviceGroup=ServiceGroup(services:[otelService,service,systemMetricsMonitor],gracefulShutdownSignals:[.sigint],cancellationSignals:[.sigterm],logger:logger)tryawaitserviceGroup.run()}}
We’re looking for contributions to grow the list of process metrics collected and to expand platform support. PRs are welcome - our contribution guidelines describe how to get started.
By reaching 1.0, this project will maintain a backwards-compatible API as it continues to evolve. Thanks to everyone who contributed to this release.
There is a very particular flavor of frustration that only appears in identity work. I used to work in the identity and network access division at Microsoft, so my joke was I needed to move faster over the course of my career, because I've landed smack dab in identity and by proxy security most of my career. The moment I'm talking about is when Microsoft Authenticator calmly informs you that you have "too many devices registered." You know with absolute clarity that this is incorrect, because...
Gavriel Cohen built NanoClaw, a lightweight alternative to OpenClaw, in a weekend after learning about security flaws in the popular agentic framework.
On this edition of The New Stack Agents, we talk to Cohen about how he is building NanoClaw, how he is using it, and what he learned about the future of programming from the experience.
Cohen is the co-founder of AI marketing agency Qwibit, where he was already running agents for everything — operations, research, sales pipeline, client management, and documentation. To do that, he used Claude Code, but he wanted to create an interface that would be friendlier to the non-technical users in the company.
“I had this idea, before starting with OpenClaw, that I’d like to have containers and agents that are running on a machine that’s always on in the background,” he says. Around the same time, Clawdbot, later renamed OpenClaw, launched and seemed to provide an answer to all of this.
“I started to run it, and then immediately it clicked that this was what I needed for both giving me and my co-founder access to the sales data and the sales pipeline in a really easy-to-use interface.”
But Cohen says he didn’t sleep well that first night after setting up Clawdbot. As he set up the project, he noted that Clawdbot had added a small GitHub package he’d created months earlier: a Gemini-based PDF editing tool with a few hundred stars and zero recent activity.
“As a developer, every single dependency that I add to my software, I vet,” Cohen says. “Being that they had added my package, which anybody who was vetting it should not have added — right away, I was like, this is worrying.”
He also noticed that the way the original Clawdbot connected to his WhatsApp account meant it didn’t just store the messages from the groups he’d told the agent to monitor; it also stored all messages in a local database.
By that point, the Clawdbot codebase had already ballooned to roughly 350,000 lines, generated in a matter of weeks with AI.
“It breaks the fundamental thing that makes open source work,” Cohen says. “The code is there, and people look at the code. But when it’s coded so quickly with so little oversight, nobody else is going to be able to audit 400,000 lines of code.”
He believes, however, that the biggest issue was architectural: there was no isolation between agents. An agent in a family WhatsApp group and one connected to a work repository ran in the same environment, separated only by application-level blocks rather than OS-level sandboxing.
That insight became the foundation for NanoClaw.
Minimal code as a design philosophy
NanoClaw launched on GitHub in late January and now has just under 10,000 stars. The core principle is radical minimalism: about a few hundred lines of actual code, a handful of dependencies, and each agent running inside its own container.
“I’m going to put just the code that I need, nothing else,” Cohen says. “Every line of code that you’re running is code that’s there for you. And not a single line of code that’s there to support someone else.”
Built on Claude Code, NanoClaw skips the typical installation wizards, configuration files, and plugin systems. Setup runs through a Claude Code skill, a Markdown instruction file that guides Claude in walking the user through the process, asking questions along the way. Apple Containers or Docker? There’s a skill file for that, too. Want to add Telegram alongside WhatsApp? Run /add-telegram, and Claude walks you through the process and builds the integration.
The entire project’s source code fits into about 35,000 tokens, roughly 17% of Claude Code’s 200,000-token context window. That means a coding agent can pull in the full codebase, understand it completely, and one-shot most features. OpenClaw’s 400,000-line codebase, by contrast, would span many context windows.
Cohen is pushing this further. The next refactor, he tells us, will strip WhatsApp out of the core and remove file-mounting code, leaving a headless runtime of about 2,000 lines. Integrations and features get added at build time through skills, so each user’s deployment contains only the code it actually runs.
“If a piece of software is adding all this functionality that you don’t need, then that software has gotten worse for you,” Cohen says. “It’s a larger package, it’s less secure — and you don’t need it.”
New rules for coding in the age of AI
Building NanoClaw reinforced something Cohen, who was previously a full-stack engineer at Wix, had been thinking about for a while: AI agents will fundamentally change how developers write and maintain code.
Take DRY — don’t repeat yourself. Cohen argues that this made sense when writing and testing code was expensive. But that has side effects when used with coding agents, because when coding agents edit a shared function, they tend to make the change and move on without ever considering the downstream effects. Duplicated code eliminates that class of side effects.
“The overhead of maintaining duplicates doesn’t cost that much anymore,” Cohen says. “You can run Claude Code on it, and it will apply the same changes throughout.”
Strict file-length linting is another area where change may be needed. Cohen says that early in his Claude Code usage, he set a 120-line maximum per file. But the result of that was that the agent spent more time refactoring to stay under the limit than building features. Today’s models can handle files of 500 to 1,000 lines with targeted edits, making the old rule counterproductive.
Cohen also argues that the value of a given piece of code is dropping fast when, every three to six months, better and cheaper models arrive. Code that works today doesn’t need to stand the test of time, he believes. In a year, a better agent will simply be able to rewrite it.
“We’re not writing code today that needs to stand the test of time for years in the future,” he says.
No code in Markdown files
One newer principle Cohen is championing: no code blocks inside Markdown skill files.
That is fine as part of Anthropic’s SKILL.md standard, but what tends to happen, he says, is that scripts sneak into instruction documents. Claude then reads the code, writes it back out to its bash tool, and executes it. That process only introduces errors, especially as the context window fills up.
The fix: Markdown files reference external scripts, and each script outputs only a few lines of status and a log file, rather than flooding Claude’s context with raw terminal output. When, for some reason, Claude needs more context, it can look at the log file.
In NanoClaw’s setup process, removing code from the Markdown file reduces token consumption to 3,000, down from 30,000 to 100,000 previously.
What’s next?
The project now has a group of regular contributors, and Cohen says the roadmap centers on shrinking the core even further and making it easier to build on top of. The goal is a runtime so small that an enterprise security team could audit it in an afternoon, whiteboard the full architecture, and verify every line.
“I hope companies will be built on top of NanoClaw,” he says. “Simple infrastructure that anybody could build on.”
As for whether Cohen himself will build one of those companies, he says the hard part isn’t finding the opportunity.
“The difficult question is, which company?” he says. “Because there’s just so many options.”
Log in
