@kalimdevio

Software Engineer

United Arab Emirates
Joined July 2018
6 Books That Helped me Remove all Negative Thoughts.
8
186
1
1,017
36,831
Kalim retweeted
I am unstoppable.
29
389
11
2,446
81,931
What's a random but important fact you think more people should know? I'll start: Most people overestimate what they can learn in a day and underestimate what they can learn in a year.
28
19
4
208
17,959
New: A map of the most important skills in AI Engineering.
512
4,099
481
24,330
6,868,056
If you don't respect small money, you will never see big money.
300
1,685
79
12,819
259,446
Just saw that the LLMs-from-scratch repository passed 100,000 stars on GitHub! This is super cool and motivating. I am really happy to see that this open-source repo has helped so many people. Thanks also to everyone who shared ideas and opened PRs with improvements! Of course, I plan to keep adding new material, including new attention variants and architectures (while bigger projects like RL and Reasoning From Scratch live in their separate repositories). I am also currently working on a larger applied custom “small” LLM project. It has been keeping me super busy this month, but I will share more on that soon. If you are new to it, some of the highlights include 1. Of course, the complete code path from tokenization and attention to pretraining, classification, and instruction fine-tuning, etc. All of it FROM SCRATCH, of course! (RL lives in a companion repo.) 2. From-scratch implementations of Llama, Qwen, Gemma, and Olmo (smaller variants that run locally and can be plugged into the training scripts). 3. From-scratch implementations of attention alternatives and other architecture components, such as GQA, MLA, sliding-window attention, Gated DeltaNet, DeepSeek Sparse Attention, cross-layer KV sharing, and mixture-of-experts 4. Materials on KV caching, training performance, memory-efficient weight loading, DPO, evaluation, and LoRA So, if you don’t have any weekend plans yet, happy tinkering!
82
556
26
3,953
171,678
Kalim retweeted
Introducing Muse Code (beta), a terminal coding agent built for long-horizon software engineering, powered by our new Muse Spark 1.2 model. Muse Code plans, implements, and validates complex, multi-file changes across large repositories with persistent sub-agents that solve difficult problems faster, more accurately, and with less intervention. 🧵👇
199
357
136
3,700
516,622
In 2022, as a Google intern, I sent a random calendar invite to Jeff Dean. He accepted. His calendar was accessible even to interns, so I figured I had nothing to lose. The next day, I sent this photo to my team. Everyone was shocked. It’s hard to imagine Google without Jeff. He’s been one of my biggest inspirations. Thank you for everything, Jeff.
Tomorrow will be my last day at Google after 27 years, and watching it grow from 25 people to 190,000+ has been an amazing journey. Below is a note I shared with many people internally at Google today. An excerpt is: It has been an absolute pleasure to work with you and to help build some of the most widely used and impactful products of all time. As a kid, I dreamed of helping build software that would be used by many people, and Google now has thirteen products used by more than a billion people (amazing!). Our work has had a tremendous impact in the world, and I have been lucky enough to collaborate and form friendships with many colleagues that I deeply admire, respect, and enjoy. It still brings me joy every time I see people out in the world using our products to find information, handle email, translate documents, watch videos, learn new things, navigate and understand the physical world, browse the web, use their phone, run large-scale computations on our infrastructure, ride in an autonomous vehicle, or perform complex tasks with the help of our AI systems. I hope you all share this sense of joy, because it is a shared accomplishment! Thank you to all of my colleagues at Google over many years! Now I'm excited to go start @DiscoLoopAI with my longtime friends and colleagues @Sanjay_Ghemawat, @OriolVinyalsML, and @quocleix. (Updated post: slightly redacted to not have some personal info)
233
703
89
35,501
9,012,146
Underrated truth: No matter where you are in life, there is still so much to look forward to. People you haven’t met. Places you haven’t seen. Ideas you haven’t discovered. Versions of yourself you haven’t become. Believe the world has more in store for you, and it will.
73
1,109
23
6,327
126,735
People fail because they do fake work. Rocking horse syndrome. It feels like you're moving forward, but in reality, you're going nowhere. Reading about business, constantly testing technology, "studying", etc. Anything but generating revenue!
157
45
9
575
26,150
Kalim retweeted
The average person does not try to solve their problems or struggles. They get stuck in a routine and live the same day over and over. It does not cross their minds to snap out of it, engage in strategic thinking, and change direction so they can eventually reach a more advantageous position in life. If you want to get ahead of 90% of people, all you have to do is pause, think, and actually attempt to overcome the struggle you're faced with.
410
1,766
134
14,871
522,111
Kalim retweeted
$31 MILLION HOTEL. EVERY PIPE AND WIRE MAPPED IN ADVANCE. BUILT BY TYPING SENTENCES INSTEAD OF PLACING THEM BY HAND. A 190-room hotel needed its ductwork, plumbing, electrical, and fire suppression modeled and coordinated before construction could start. The design team connected Kimi K3 to their modeling software through MCP, then had it read photos of the blueprints alongside plain English descriptions of each system and build the model directly instead of placing every component by hand. Old process: 3 engineers, 6 weeks, around $47,000. New process: 1 engineer confirming the AI's output, 9 days, roughly $10,500 combined, engineer's time plus AI compute. Coordinated models finished this way cut change orders by 60 to 80 percent once construction begins. See the article below for a smaller example of the Kimi K3 + Blender MCP workflow.
Readers added context they thought people might want to know
Não há nenhuma reportagem independente, anúncio de empresa de engenharia/construção, nome do hotel/projeto, localização, ou documentação pública confirmando esse caso específico de US$ 31 milhões. x.com/i/grok/share/3…
245
1,041
146
9,496
1,653,823
Kalim retweeted
We quietly released the open-source Codex Security CLI, but Hacker News found it before we had a chance to share it here... You can now use it to scan repositories, track findings across runs, verify fixes, and add security checks to CI/CD. This is an early release, and we're listening to your feedback as we continue improving it.
455
1,201
241
13,550
1,365,959
Kalim retweeted
GeoLibre v2.3.0 is here! GeoLibre is a free and open-source, lightweight, cloud-native GIS platform for visualizing, exploring, and analyzing geospatial data. It runs everywhere you do, in the web browser, on the desktop, on mobile, and inside Jupyter notebooks, all while keeping your data local and private. This release brings a legend that writes itself from your symbology, a new GeoLens catalog browser, and 200+ GeoLibre Rust geoprocessing tools running entirely in the browser. What's new in v2.3.0 - Automatic on-map Legend: the legend builds itself from your visible layers, with class rows for graduated, categorized, rule-based, and expression styling, gradient bars for heatmaps and raster colormaps, and land-cover labels from a Raster Attribute Table. Rename, hide, reorder, or add your own entries, and it saves with the project. - Symbology swatches in the Layers panel: every row shows a dot, line, square, or image glyph in the layer's own color, so a tall layer stack reads at a glance. - GeoLens catalog browser: connect to a self-hosted GeoLens server, search its catalog, and add datasets as vector tiles, GeoJSON, or rendered raster tiles. - Emerging Hot Spot Analysis: build a space-time cube from timestamped points and classify every cell as a new, intensifying, persistent, diminishing, sporadic, oscillating, or historical hot or cold spot, all client side. - Mosaic time series: the Time Slider now steps through MosaicJSON and STAC collections of many COGs per date, on either a GPU or a WASM rendering engine. - Copy and paste layer styles: give a whole set of layers one consistent look without restyling each in turn. - Shareable tool links: deep-link any Whitebox tool with a ?tool= URL that opens the dialog preselected and pre-fills the form, with a Copy link button to build it for you. - Smarter data loading: pick which layers to load from a multi-layer GeoPackage, import CSVs whose coordinates are in any projected CRS, and read a raster's real CRS, pixel size, and extent from the metadata dialog. - Multiple AI profiles: define several provider, model, and credential setups, pick a default, and switch between them from the assistant panel. Try it out - Launch GeoLibre Web: web.geolibre.app - GitHub: github.com/opengeos/GeoLibre - Documentation: geolibre.app - Release notes: github.com/opengeos/GeoLibre… #GIS #Geospatial #OpenSource #RemoteSensing #MapLibre #GeoLibre
62
726
36
4,533
294,600
There are only five levels of income: Level 1 - you work Level 2 - your team works Level 3 - your systems work Level 4 - your product works Level 5 - your money works
142
572
19
4,698
204,297
Kalim retweeted
Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use. Its capabilities exceed those of any model we’ve ever made generally available.
4,939
14,187
9,951
103,766
57,618,079
A French engineer who lives quietly in Paris has spent 30 years writing software that the entire internet now runs on without knowing his name. He wrote the code that streams every YouTube video, every Netflix show, every TikTok clip. He wrote the code that runs the virtual servers underneath AWS, Google Cloud, and Microsoft Azure. He calculated more digits of pi than anyone in history. He has no Twitter. He has no marketing. He just keeps shipping. His name is Fabrice Bellard. Here is the story, because almost nobody outside the systems programming world knows what one man has built. Fabrice was born in 1972 in Grenoble, France. He studied at École Polytechnique, the top French engineering school. He never went to Silicon Valley. He never built a startup empire. He just wrote code. In 2000 he started a project called FFmpeg, an open-source multimedia framework for encoding, decoding, and streaming video. He was 28. The project did one thing nobody else had done well. It handled every video and audio format that existed, in one library, on every operating system. He led it himself for years. Today FFmpeg is the invisible engine of the internet. YouTube uses it. Netflix uses it. VLC uses it. Chrome and Firefox use parts of it. Every Android phone, every iPhone, every smart TV, every video editing tool you have ever touched runs FFmpeg somewhere underneath. If you have watched a video on a screen in the last 20 years, Fabrice's code processed it. He was not done. In 2003 he started QEMU, a machine emulator and virtualizer. He wrote it solo until version 0.7.1 in 2005. QEMU lets you run any operating system on any other operating system. It became the foundation of modern virtualization. KVM, the Linux kernel hypervisor, runs on top of QEMU. Every major cloud provider, AWS, Google Cloud, Microsoft Azure, IBM Cloud, runs virtual machines on infrastructure built around it. The Quick Emulator is the most cited piece of cloud infrastructure code on Earth. He kept going. In 2001 he won the International Obfuscated C Code Contest with a small C compiler that grew into TCC, the Tiny C Compiler. TCC can compile and boot a Linux kernel from source in under 15 seconds. In 2004 he calculated the most digits of pi ever computed at the time, using a personal desktop computer and an algorithm he derived himself called Bellard's formula. In 2011 he wrote a complete PC emulator in pure JavaScript that runs Linux in your browser, a project called JSLinux that engineers still cannot believe is real. In 2019 he released QuickJS, a small but complete JavaScript engine that fits where V8 cannot. In 2021 he released NNCP, a neural network based lossless data compressor that immediately took the lead on the Large Text Compression Benchmark. Then he turned his attention to large language models. He built TextSynth Server, a web server with a REST API for running LLMs locally. He released ts_zip and ts_sms, compression utilities that use language models to compress text and short messages at ratios traditional algorithms cannot reach. He released TSAC, a very low bitrate audio compression system. In December 2025 he released Micro QuickJS, a new JavaScript engine for microcontrollers, separate from QuickJS, designed for environments with almost no memory. Fabrice co-founded a telecom company called Amarisoft in 2012, where he serves as CTO. Amarisoft builds 4G and 5G base station software used by carriers and labs around the world. He has been running it for over a decade while continuing to ship personal projects from his own home page at bellard dot org He has no Twitter. He has no Instagram. He gives almost no interviews. His personal website is a flat list of projects with no styling, no fonts, no marketing copy. Just titles and links. A quiet French engineer who never moved to Silicon Valley wrote the code that quietly runs the internet. He is still shipping.
380
4,502
404
25,376
3,192,581
Kalim retweeted
It's not FAANG anymore. It's MANGO.
632
1,596
364
25,499
4,282,521
Kalim retweeted
Introducing the OpenAI Deployment Company, which will help businesses maximally succeed with their deployments of AI. Starting with 150 Forward Deployed Engineers and Deployment Specialists, and $4 billion of initial investment from 19 partners.
Today we’re launching the OpenAI Deployment Company to help businesses build and deploy AI. It's majority-owned and controlled by OpenAI. It brings together 19 leading investment firms, consultancies, and system integrators to help organizations deploy frontier AI to production for business impact. openai.com/index/openai-laun…
267
144
57
2,616
684,191
“I don’t think I’ve typed a line of code since December.” When Andrej Karpathy said that, most people treated it like a crazy AI quote. @garrytan treated it like a question: “What happens when one person operates like an entire software team?” Then he built gstack. And honestly… this repo feels less like a dev tool and more like a preview of where software is going. Not AI as autocomplete. AI as: - CEO - Staff engineer - QA lead - Security reviewer - Designer - Release manager - Browser operator - Parallel execution layer All coordinated through structured workflows. The craziest part is the numbers. Garry says his current pace is ~810× higher than his 2013 output — normalized for logical code changes, not inflated AI LOC. Same person. Same brain. Different tooling. That’s the shift everyone is underestimating right now. The winners in the next era probably won’t be the people who code the fastest. They’ll be the people who can direct, review, and orchestrate AI systems the best. A few things in gstack that genuinely stood out to me: → /office-hours challenges your product assumptions before you build → /autoplan runs CEO + design + eng reviews automatically → /qa opens a real browser, tests flows, finds bugs, and fixes them → /review catches production-level issues before shipping → /pair-agent lets multiple AI agents collaborate together → parallel AI sprints running at the same time across projects This is the first open-source repo in a while that actually made me stop and rethink how software teams will work 2–3 years from now. We’re moving from: “AI helps developers code” to “developers operate systems of AI workers.” That’s a very different future. 100% Open-source Link in comments 👇
31
122
6
673
259,387