Victor Morales Ruvalcaba

TerminalTelemetry v0.12 - Major Release is out! PyQt6 for Network Tools: Why I Ditched WebEngine for Native Widgets Everyone builds network automation tools as web apps or CLIs. I've been delivering different tools for the desktop - and found Qt's native widgets are the amazing framework nobody talks about. If you're familiar with KDE Desktop on Linux, you've used Qt. It's powerful - Qt Designer lets you build GUIs visually. pyuic6 generates your Python GUI code, then you wire in your logic. Draw your UI, generate the code, and off you go. In some ways it's like Visual Basic for Python! (for those who get that reference). Visual Basic opened the door to Windows programming for so many in my generation. It was huge. The mistake: TerminalTelemetry v1 used Qt WebEngine. Seemed like a good idea - enabled some unique UI rendering, web tech in a desktop wrapper, right? The reality: Memory bloat, slower performance, and threading headaches. Plus all the complexity of web development without the benefits. Hybrid WebEngine apps have their place (that's why you see WebView in C# and JavaFX), but Qt's event-driven architecture with Signals and Slots was perfect for my new widget design. The fix: v0.12 moved to pure PyQt6 native widgets and native threading. What I gained: - Native performance (no browser overhead) - Real threading (SSH operations don't freeze the UI) - Rich widgets (tables, trees, progress bars that actually work) - Cross-platform (Windows, macOS, Linux) - Memory efficiency - Full integration of all the session manager and credentials manager functionality! What TerminalTelemetry delivers: SSH-only monitoring with real-time telemetry: - Live AND exportable CDP/LLDP neighbors, routing tables, ARP entries - TextFSM templates you can customize right in the GUI - Extensible multi-vendor support - switched from Napalm to pure Netmiko + templates The JSON-based platform system sets me up to support any platform the excellent Netmiko supports. Thanks again Kirk Byers! v0.12 also includes: automatic theme downloading, custom theme editor (make your own themes!), and multiple bug fixes. Really just a better product now. As always, here is the code: https://lnkd.in/gFa3Xvcs And in case you haven't thought about it, Secure Cartography, my Python based mapping solution is also a PyQt6 app. Check it out here: https://lnkd.in/gE3iZZBM

72

3 Comments

Migrations are the least fun - and most challenging - part of scaling. Scaling in the age of AI isn’t just about running a lean team - it’s also about tackling migrations, managing compute costs, and building products that can handle millions of users without breaking. Michael T. (of AnySphere/Cursor fame) and Parker Conrad (of Rippling fame 😃) recently sat down to discuss a wide range of topics: ⚙️ Why migrations are the hardest (and least fun) part of scaling ⚙️ How AI changes the calculus of team size and productivity ⚙️ The trade-offs between speed, distribution, and long-term margin ⚙️ Building technical moats in markets moving at AI speed 👉 If you’re building in AI right now, or thinking of building something soon, give this a watch! https://lnkd.in/ggpMtcyG

274

7 Comments

If you want to become good at system design (in 1 month or less), learn these case studies: 1 How Cloudflare Supports 55 Million Requests per Second With Only 15 Postgres Clusters ↳ https://lnkd.in/eEQP6Apw 2 How PayPal Was Able to Support a Billion Transactions per Day With Only 8 Virtual Machines ↳ https://lnkd.in/eqcb7MpP 3 How YouTube Was Able to Support 2.49 Billion Users With MySQL ↳ https://lnkd.in/efmJw4Dx 4 How Google Ads Was Able to Support 4.77 Billion Users With a SQL Database ↳ https://lnkd.in/efnSvwJp 5 Real-Time Gaming Leaderboard Architecture ↳ https://lnkd.in/edfTDq5R 6 How Stripe Prevents Double Payment Using Idempotent API ↳ https://lnkd.in/erMkqwq4 7 How Amazon S3 Achieves 99.999999999% Durability ↳ https://lnkd.in/eutGiK35 8 How Uber Finds Nearby Drivers at 1 Million Requests per Second ↳ https://lnkd.in/eeqH9Hjh 9 How to Scale an App to 10 Million Users on AWS ↳ https://lnkd.in/eU736g9Q 10 How Slack Works ↳ https://lnkd.in/eATMDjrK 11 How Distributed Counter Works ↳ https://lnkd.in/eGwaA62J 12 How Pastebin Works ↳ https://lnkd.in/eZpfaVjc 13 How Amazon S3 Works ↳ https://lnkd.in/e2p7qXri 14 How Uber Computes ETA at 0.5 Million Requests per Second ↳ https://lnkd.in/eVKV2ePC What else should make this list? —— 👋 PS - Join 100,000 engineers and get the powerful system design template (it's free): → https://lnkd.in/eWAp5ZwE ——— ♻ Repost to help others learn system design. ➕ Follow Neo Kim for more.

1,182

85 Comments

Surprising details about the Grok-3 release: * Their current cluster already has 200k H100s/H200s. They initially reached out to datacenter facilities to ask how long for them to build them a 100k cluster and the timelines were 12-24 months, which was too slow, Elon said they'd definitely lose if they went that route. So they found an abandoned factory in Memphis, an empty shell, and built custom electrical / cooling systems, using portable generators, Tesla packs to smoothen out the power spikes (due to the fact they're using synchronous gradient updates), etc. It took them ~122 days to build the whole thing e2e with 100k H100s. And additional ~90 days to add 100k more. No one has ever done something like this. * Elon announced they're building a new ~1.2 GW cluster of GB200s/GB300s - this is OOM larger than any other datacenter in the world, and their current datacenter is already the largest single cluster in the world. * Igor said that while they were running Grok-3, AI engineers used to go to the cluster physically and plug off a node to make sure the run is robust to such pertrubations -> this is one of the things Elon does great, reduces barriers between designers/engineers, engineers/datacenter technicians, etc. * Grok-3 is the first model to pass 1400 score on the arena :O The scary obvious thing here is that due to the culture the team has, Elon's capability to attract capital and talent, and the rate of progress, I don't think anyone will be able to compete with them. * They said they'll open-source Grok-2 as soon as Grok-3 is stable, in a few months. And they plan to keep that strategy going forward, open-sourcing last generation but still managing to stay competitive. They hide the chain of thought the same way OpenAI did. Original video here: https://lnkd.in/dEHKE6BV

401

48 Comments

Amazon ECS now supports Managed Instances. When enabled for an ECS cluster, you get automatic selection of the most cost-optimized EC2 instance type, with the choice based on specific attributes (memory, CPU manufacturer, vCPU count, GPU architecture, and much more) or on the instance type. As part of this feature, ECS takes full responsibility for infrastructure management including provisioning, scaling, maintenance, and patching. It also manages resource utilization, placing multiple tasks on large instances, consolidating workloads, and terminating idle instances. Read "Announcing Amazon ECS Managed Instances for containerized applications" by Micah Walter at https://lnkd.in/gwyGXJDJ .

507

14 Comments