0nel Arrieta retweeted
Every serious cloud workload whether it is simple VM setup, an AKS cluster, or a massive hybrid architecture , starts with intentional network design.
Get the VNet right first, or spend months debugging routing issues and security holes later.
Here is a complete structural breakdown of Azure Virtual Networks, from CIDR boundaries to hybrid connectivity paths.
👇 Save this cheat sheet for your next architecture design review.
You heard "cloud computing" and thought someone else manages everything now.
Not quite.
If you are running Azure VMs or AWS EC2, you are still a system administrator. Microsoft and AWS just own the physical rack.
The architectural differences every engineer needs to know:
1/ Networking Topology
AWS attaches stateful Security Groups directly to instances. Azure separates the NIC as a standalone resource, letting you apply Network Security Groups at both the subnet and individual NIC level. More granular. More control.
2/ The Ephemeral Trap
Azure VMs include a local temporary disk by default. If your VM is stopped and deallocated, that data is wiped instantly. It is for swap space and caching only. Beginners lose production data here constantly.
3/ The Enterprise Math
Azure Hybrid Benefit is the cost differentiator AWS has no answer to. If your enterprise already owns on-prem Windows Server or SQL Server licenses, you port them to Azure and cut compute costs significantly. One feature alone shifts the cloud decision for most Microsoft shops.
Virtual Machines give you total control. But control means responsibility.
If you want the provider to handle patching, runtime updates, and auto-scaling, move up the stack to PaaS or Serverless.
Understand what you are managing. That clarity is what separates engineers who just use cloud services from engineers who actually architect them.
#Azure
0nel Arrieta retweeted
It continues to be a fascinating time to work in #DistributedSystems! Here's a reading list of 70+ Distributed Systems papers, *all from 2025*!
CS525 UIUC Spring 2026 Reading List (new!): go.siebelschool.illinois.edu…
(Slides available in "Schedule" in above doc on those dates).
It's STILL a fascinating time to be working in Distributed Systems! Here's a reading list of 70+ Distributed Systems papers, *just from 2024*!
CS525 UIUC Spring 2025 Reading List (new!): go.siebelschool.illinois.edu…
(Slides in "Schedule" in above doc, available on those dates).
0nel Arrieta retweeted
CMU DB Group is brilliant, and hosts some of the best seminar series. This one nails modern cloud DB metadata, branching, and GC in a super approachable way.
db.cs.cmu.edu/events/futured…
0nel Arrieta retweeted
Red team platform with LLM agent and automation workflows
github.com/FunnyWolf/Viper
0nel Arrieta retweeted
we’re less than 10 years from decoding all animal communication and BCIs that let us talk back
an entire kingdom of minds we’ve been ignoring for 300,000 years
going to be really fun
0nel Arrieta retweeted
CVE-2026-21858 + CVE-2025-68613: n8n Ni8mare - Full Chain Exploit
Unauthenticated to Root RCE:
- LFI via Content-Type confusion
- Read /proc/self/environ to find HOME
- Steal encryption key + database
- Forge admin JWT token
- Expression injection sandbox bypass
- RCE as root
CVSS 10.0
github.com/Chocapikk/CVE-202…
0nel Arrieta retweeted
UC Berkeley's "Machine Learning" lecture notes
PDF: people.eecs.berkeley.edu/~jr…
0nel Arrieta retweeted
List of courses with certificates and badges
github.com/PanXProject/aweso…
0nel Arrieta retweeted
Download a Movie Get a RAT: youtu.be/J9G66yojDUo
0nel Arrieta retweeted
1. Google Dataset Search
Aggregates datasets from across the web
Find niche data others miss
datasetsearch.research.googl…
2. Data.gov
250K+ government datasets
Real policy and economic data
data.gov/
0nel Arrieta retweeted
UCI Machine Learning Repository
600+ datasets used in actual research papers
Battle-tested, well-documented
archive.ics.uci.edu/
0nel Arrieta retweeted
We're donating MCP to the @linuxfoundation and launching the Agentic AI Foundation with @OpenAI, @blocks, @AWS, @Bloomberg, @Cloudflare, @Google, and @Microsoft.
MCP went from internal project to industry standard in a year. Now it gets the long-term stewardship it deserves.
Anthropic is donating the Model Context Protocol to the Agentic AI Foundation, a directed fund under the Linux Foundation.
In one year, MCP has become a foundational protocol for agentic AI. Joining AAIF ensures MCP remains open and community-driven. anthropic.com/news/donating-…
0nel Arrieta retweeted
#reversing
#Tech_book
#Kernel_Security
"Reverse Engineering Armv8-A Systems:
A practical guide to kernel, firmware, and TrustZone analysis", 2025
]-> Repo - github.com/PacktPublishing/R…
// Written by a Linux kernel engineer and author with over 14 years of industry experience, this book lays a solid foundation in reverse engineering and takes you from curious analyst to expert
0nel Arrieta retweeted
I recently came across an open-source project called AWS FinOps Dashboard, and it does something very practical that most teams struggle with: giving engineers a fast, terminal-friendly view of cloud spend.
It connects to one or many AWS accounts and pulls cost, usage, and resource data directly into the CLI. That makes it easy to see which services drive the bill, how costs change over time, and which tags or environments are responsible.
It also surfaces untagged resources, idle instances, and patterns that usually get ignored until budgets spike. For EC2-heavy setups, even a simple view of stopped or underused instances can guide quick fixes that save money immediately.
Because it’s all open source, the workflow feels natural. You can export reports, integrate it into scripts, or run periodic audits without relying on a full FinOps platform.
Tools like this won’t replace enterprise dashboards, but they make cost visibility part of everyday engineering work. And that’s usually where meaningful optimisation starts.
0nel Arrieta retweeted
Today we release the transformers version 5 RC! 🤗
With this, we enable e2e interoperability with our friends in ecosystem, ease up adding new models and simplify the library 🙌🏻
Read our blog to learn more: huggingface.co/blog/transfor…
0nel Arrieta retweeted
Today's Future Data Systems Seminar Speaker: Prashant Singh from @Snowflake will present the @ApachePolaris re-implementation of the @ApacheIceberg REST catalog API. Zoom talk open to public at 4:30pm ET. YouTube video available after: db.cs.cmu.edu/events/futured…
0nel Arrieta retweeted
The Path of a Packet Through the Linux Kernel
net.in.tum.de/fileadmin/TUM/…
0nel Arrieta retweeted
This article describes how the author scaled a FastAPI service to handle 200,000 requests per minute using Gunicorn + Uvicorn workers and Kubernetes Horizontal Pod Autoscaler, with async endpoint optimization and autoscaling strategies
➤ ku.bz/k7xBpkm1J
0nel Arrieta retweeted
Airbus is rolling out a critical software update. Around 6,000 A320 aircraft have been grounded. The reason: solar radiation can cause failures in the onboard computer.
Recently, an A320 experienced an uncontrolled down “pull” while the autopilot was engaged. The cause turned out to be a problematic software update that led to a failure in the flight control computer (ELAC). In the worst-case scenario, this could push the aircraft beyond its structural limits. The ELAC system is designed with redundancy. Two onboard computers cross-check each other to avoid errors. When one provides incorrect data, the other should detect it and take over.
In software version L104, this logic was faulty, it failed to detect corrupted data caused by cosmic radiation. When radiation “flipped” bits in memory (e.g. from 0 to 1 or from 1 to 0, which does happen), the system did not recognize the error and executed an incorrect command. The solution is to revert to an older software version





































