nietras 👾
@nietras1
Systems and machine learning programmer pushing .NET performance to the limits for real-time 24/7 edge applications. Author of https://github.com/nietras/Sep
Updated "Sep 0.10.0 - 21 GB/s CSV Parsing Using SIMD on AMD 9950X 🚀" to make it 300% clear the graph shows #performance progression over different Sep, #dotnet versions and CPUs. To show how runtime and library improvements go hand in hand with hardware changes. Eg AVX512 #SIMD
New blog post "Sep 0.10.0 - 21 GB/s CSV Parsing Using SIMD on AMD 9950X 🚀" 📈 Sep #performance from 7 GB/s to 21 GB/s over last two years 🧑💻 #csharp #SIMD and #x64 assembly on #dotnet 9.0 🛠️ Tweaks and new #AVX512-to-256 parser 🔢 Lots of benchmarks 👇 nietras.com/2025/05/09/sep…
RK3688 update: 8xA730+4xA530, 2TFlops GPU, 32TOPS NPU, LPDDR6 200GB/s, UCIe and more... #RKDC25 Radxa ROCK 6 coming
My friend @vinogradovk is launching something new — an Open Source Endowment (OSE), a US-based nonprofit that will fund open-source work that silently powers the world. I've joined in. Maybe you should, too 🤷🏻 Over the last 12 months, I have reached out to several tech companies…
We ran a randomized controlled trial to see how much AI coding tools speed up experienced open-source developers. The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't.
As a 31yo Principal Engineer with 20 years of programming experience and 15 years in the industry, I I would add that you should know Vacuum fluctuations Planck-scale quantum foam Feynman path integrals Quantum electrodynamics Quantum chromodynamics The Standard-Model Lagrangian…
As a Senior Software Engineer with 10+ years of experience, you should know: MySQL PostgreSQL SQLite MongoDB DynamoDB Cassandra LevelDB RabbitMQ Kafka Redis Docker K8S Nix Linux C Python Bash JS HTML CSS JIRA
We merged a C# 14 feature into .NET 10 preview 7: extension operators. We'd previously merged extension methods and properties (static and instance) but I forgot to tweet that. Example and spec below.
Deep learning training is a mathematical dumpster fire. But it turns out that if you *fix* the math, everything kinda just works…fp8 training, hyperparameter transfer, training stability, and more. [1/n]
It sure is nice that #dotnet #nuget forces a vulnerability check that completely prevents building a project due to restore failing: dotnet restore --interactive warning NU1900: Error occurred while getting package vulnerability data: Unable to load the service index for source
🎉 Big news! We've secured a $3M, three-year sponsorship from @DevolutionsInc This sponsorship will allow us to speed up development, improve docs & tooling for all the community's benefit! 🚀 Read about it here: github.com/AvaloniaUI/Ava…
Another classic has made it into the Transformers library: LightGlue (ICCV '23)🔥 A deep neural network that learns to match local features across images. Faster and more efficient than SuperGlue. Adaptive computation based on difficulty🕵️ Now available in a few lines of code!
Ever wonder why switching optimizers never actually helps? Introducing HeavyBall Benchmark
New blog post "Sep 0.11.0 - 9.5 GB/s CSV Parsing Using ARM NEON SIMD on Apple M1 🚀" 🛠️ New #ARM #NEON #SIMD parser based on @geofflangdale bulk move mask 📈 Sep #performance up from 7 GB/s on #Apple #M1 🧑💻 #csharp and #ARM assembly on #dotnet 9.0 Link below or via profile 👇
#CVPR2025 Paper Picks #1 VisionZip is a compression technique that reduces number of visual tokens to improve performance AND prefill time for vision language models model and demo on Hub, keep reading for links & paper explanation
#dotnet 10 Preview 5 released 🚀 Runtime: 👉 Escape Analysis for Delegate 👉 Inlining Improvements 👉 ARM64 Write Barrier Improvements C#: 👉 User defined compound assignment operators github.com/dotnet/announc…
A monumental collaborative effort with many incredible people ☺️ Proud to be part of this! arxiv.org/abs/2506.06536