Hacker News headlines

1

GEMM with Thread Block Clusters on Nvidia Blackwell GPUs (colfax-intl.com)

3 days ago | ashvardanian | colfax-intl.com | newest

2

Visualizing Foursquare Places with ClickHouse (clickhouse.com)

4 days ago | ashvardanian | clickhouse.com | frontpage

1

Meta Perception Encoder (meta.com)

a week ago | ashvardanian | meta.com | newest

32

Load-Store Conflicts (zeux.io)

a week ago | ashvardanian | zeux.io | best

3

Vibe Coding Is Rapidly Reshaping the Software Developer Profession (thenewstack.io)

2 weeks ago | ashvardanian | thenewstack.io | newest

52

CubeCL: GPU Kernels in Rust for CUDA, ROCm, and WGPU (github.com/tracel-ai)

2 weeks ago | ashvardanian | github.com | best

2

Burn v0.17: Deep Learning in Rust gets new backends and improved kernel fusion (github.com/tracel-ai)

2 weeks ago | ashvardanian | github.com | newest

1

Cutlass Tutorial: Writing GEMM Kernels Using Tensor Memory for Blackwell GPUs (colfax-intl.com)

3 weeks ago | ashvardanian | colfax-intl.com | newest

84

Less Slow C++ (github.com/ashvardanian)

3 weeks ago | ashvardanian | github.com | best

1

Calling CUDA in 3000 Words (ashvardanian.com)

a month ago | ashvardanian | ashvardanian.com | newest

2

Redpanda debuts new Agentic AI platform with MCP (redpanda.com)

a month ago | ashvardanian | redpanda.com | newest

3

The Tragic Case of Intel AI (geohot.github.io)

a month ago | ashvardanian | github.io | frontpage

1

Arm Introduces New Developer Initiative to Expedite Migration on Cloud Platforms (arm.com)

a month ago | ashvardanian | arm.com | newest

4

Show HN: Less Slow C++: Revisiting Performance Tricks for C/C++/CUDA/Asm/PTX (github.com/ashvardanian)

a month ago | ashvardanian | github.com | newest

3

Isomorphic Labs announces $600M external investment round (isomorphiclabs.com)

a month ago | ashvardanian | isomorphiclabs.com | newest

2

Mixing ARM NEON with SVE code for fun and profit (lemire.me)

a month ago | ashvardanian | lemire.me | newest

4

CMake 4.0.0 Released (kitware.com)

a month ago | ashvardanian | kitware.com | frontpage

1

Unlock High-Precision Keyword Search with pinecone-sparse-English-v0 (pinecone.io)

a month ago | ashvardanian | pinecone.io | newest

1

Democratizing AI Compute, Part 7: What about Triton and Python EDSLs? (modular.com)

a month ago | ashvardanian | modular.com | newest

42

Nvidia Dynamo: A Datacenter Scale Distributed Inference Serving Framework (github.com/ai-dynamo)

a month ago | ashvardanian | github.com | best

1

Git 2.49 Released with Faster Packing, Rust Foreign Language Interface (phoronix.com)

a month ago | ashvardanian | phoronix.com | newest

2

Cerebras Announces Six New AI Datacenters Across North America and Europe (cerebras.ai)

2 months ago | ashvardanian | cerebras.ai | newest

1

`std::generator`: Standard Library Coroutine Support (microsoft.com)

3 months ago | ashvardanian | microsoft.com | newest

2

Deep Dive into Matrix Optimization on AMD GPUs (seb-v.github.io)

3 months ago | ashvardanian | github.io | newest

3

The Longest Nvidia PTX Instruction (ashvardanian.com)

3 months ago | ashvardanian | ashvardanian.com | frontpage

2

CPU Ports and Latency Hiding on x86 (ashvardanian.com)

4 months ago | ashvardanian | ashvardanian.com | newest

2

Tenstorrent Wormhole Series Part 7: Bits of the MatMul (corsix.org)

4 months ago | ashvardanian | corsix.org | newest

2

GitHub Models (github.com/marketplace)

4 months ago | ashvardanian | github.com | newest

1

Camera Calibration: What to perfect before touching the code (nikolasent.github.io)

4 months ago | ashvardanian | github.io | newest

10

Nvidia Statement on the Biden Administration's Misguided 'AI Diffusion' Rule (nvidia.com)

4 months ago | ashvardanian | nvidia.com | frontpage

3

Biden unveils last round of AI chip curbs aimed at China, Russia (cnn.com)

4 months ago | ashvardanian | cnn.com | newest

1

Parsing JSON in C and C++: Singleton Tax (ashvardanian.com)

4 months ago | ashvardanian | ashvardanian.com | newest

14

Volodymyr Zelenskyy on Lex Fridman Podcast #456 (youtube.com)

4 months ago | ashvardanian | youtube.com | frontpage

2

Unifying Generative and Dense Retrieval for Sequential Recommendation (arxiv.org)

5 months ago | ashvardanian | arxiv.org | frontpage

2

Efficient In-Place UTF-16 Unicode Correction with ARM Neon (lemire.me)

5 months ago | ashvardanian | lemire.me | newest

4

Scaling data collection for training software engineering agents (nebius.com)

5 months ago | ashvardanian | nebius.com | newest

5

Linux on GPU (dmitryduka.github.io)

5 months ago | ashvardanian | github.io | newest

1

Amazon's AI Self Sufficiency – Trainium2 Architecture and Networking (semianalysis.com)

5 months ago | ashvardanian | semianalysis.com | newest

2

Outperforming cuBLAS on H100: A Worklog (cudaforfun.substack.com)

6 months ago | ashvardanian | substack.com | newest

2

Sign operation using VFIXUPIMM in AVX-512 (wunkolo.github.io)

6 months ago | ashvardanian | github.io | newest

2

Cerebras 748x faster than Frontier supercomputer in molecular simulation (networkworld.com)

6 months ago | ashvardanian | networkworld.com | newest

1

The Next 31 Years of Developing Unum (ashvardanian.com)

6 months ago | ashvardanian | ashvardanian.com | newest

2

Understanding SIMD: Infinite Complexity of Trivial Problems (modular.com)

6 months ago | ashvardanian | modular.com | newest

1

SIMD Library for Evaluating Elementary Functions, Vectorized Libm and DFT (github.com/shibatch)

6 months ago | ashvardanian | github.com | newest

1

The Microsoft Azure HBv5 and AMD MI300C (servethehome.com)

6 months ago | ashvardanian | servethehome.com | newest

1

Azure HBv5: 352 AMD Genoa Cores with over 400 GB of HBM3 (microsoft.com)

6 months ago | ashvardanian | microsoft.com | newest

11

Waiting for many things at once with io_uring (mazzo.li)

6 months ago | ashvardanian | mazzo.li | best

2

Is Prefix of String in Table? A Journey into SIMD String Processing (trent.me)

6 months ago | ashvardanian | trent.me | newest

1

iMac Announcement – October 28 [video] (youtube.com)

7 months ago | ashvardanian | youtube.com | newest

2

Create Index Externally: Offloading Pgvector Indexing from Postgres (lantern.dev)

7 months ago | ashvardanian | lantern.dev | newest

2

Tenstorrent Wormhole Series Part 6: Vector instruction set (corsix.org)

7 months ago | ashvardanian | corsix.org | newest

4

Over-Engineering 5x Faster Set Intersections in SVE2, AVX-512, & Neon (ashvardanian.com)

8 months ago | ashvardanian | ashvardanian.com | newest

1

Grounding AI in reality with a little help from Data Commons (research.google)

8 months ago | ashvardanian | research.google | newest

2

DataGemma: Using real-world data to address AI hallucinations (blog.google)

8 months ago | ashvardanian | blog.google | newest

1

Holy Macroni A recipe for progressive language enhancement (2023) (trailofbits.com)

8 months ago | ashvardanian | trailofbits.com | newest

4

Chai-1: Decoding the molecular interactions of life (chaidiscovery.com)

8 months ago | ashvardanian | chaidiscovery.com | newest

1

35% Discount on Keyword Arguments in Python (ashvardanian.com)

8 months ago | ashvardanian | ashvardanian.com | newest

4

Cerebras reaches 1800 tokens/s for 8B Llama3.1 (forbes.com/sites/craigsmith)

9 months ago | ashvardanian | forbes.com | frontpage

1

The Painful Pitfalls of C++ STL Strings (ashvardanian.com)

9 months ago | ashvardanian | ashvardanian.com | newest

1

Stack-PR: an open source tool for managing stacked PRs on GitHub (modular.com)

10 months ago | ashvardanian | modular.com | newest

6

Benchmarking ARM Processors: Graviton 4, Graviton 3 and Apple M2 (lemire.me)

10 months ago | ashvardanian | lemire.me | frontpage

2

FAIR's Chameleon Early Fusion Code (github.com/facebookresearch)

11 months ago | ashvardanian | github.com | newest

1

We Found Something Testing the 5th Gen Intel Xeon Scalable 1P CPUs (servethehome.com)

12 months ago | ashvardanian | servethehome.com | newest

2

Gemma-10M Technical Overview (medium.com/akshgarg_36829)

a year ago | ashvardanian | medium.com | newest

1

How fast can construct a small list of strings in C for Python? (lemire.me)

a year ago | ashvardanian | lemire.me | newest

2

Meilisearch 1.8 (meilisearch.com)

a year ago | ashvardanian | meilisearch.com | newest

1

Visual Language Models on Nvidia Hardware with VILA (nvidia.com)

a year ago | ashvardanian | nvidia.com | newest

5

Show HN: Swift-powered AI apps on iOS: Real-time multimodal semantic search (github.com/ashvardanian)

a year ago | ashvardanian | github.com | newest

2

Liburing 2.6 Released (github.com/axboe)

a year ago | ashvardanian | github.com | frontpage

1

MinSH: Near-linear time global string alignment in a few lines of Python (github.com/pesho-ivanov)

a year ago | ashvardanian | github.com | newest

1

FM-Indexes and Backwards Search (2011) (alexbowe.com)

a year ago | ashvardanian | alexbowe.com | newest

56

Ollama v0.1.33 with Llama 3, Phi 3, and Qwen 110B (github.com/ollama)

a year ago | ashvardanian | github.com | best

1

GCC 14.1 Compiler Aiming for Release Around 7 May with C23 and AVX10.1 (phoronix.com)

a year ago | ashvardanian | phoronix.com | frontpage

2

Writing a WASM Runtime in Rust (skanehira.github.io)

a year ago | ashvardanian | github.io | newest

1

Multimodal Embeddings for JavaScript, Swift, and Python (github.com/unum-cloud)

a year ago | ashvardanian | github.com | newest

1

PyTorch internals: ezyang's blog (2019) (ezyang.com)

a year ago | ashvardanian | ezyang.com | newest

4

KraftCloud (unikraft.io)

a year ago | ashvardanian | unikraft.io | newest

40

KraftCloud (github.com/unikraft)

a year ago | ashvardanian | github.com | best

2

Show HN: UForm v2 Featuring Multimodal Matryoshka, Multimodal DPO, and ONNX (github.com/unum-cloud)

a year ago | ashvardanian | github.com | newest

1

VideoPrism: A foundational visual encoder for video understanding (research.google)

a year ago | ashvardanian | research.google | newest

7

Technical Deep Dive: AI Accent Localization for Call Centers (krisp.ai)

a year ago | ashvardanian | krisp.ai | newest

2

Tenstorrent Kernel by George Hotz (github.com/geohot)

a year ago | ashvardanian | github.com | frontpage

3

LLM Inference Speed of Light (zeux.io)

a year ago | ashvardanian | zeux.io | frontpage

4

C++ meets TypeScript: bidirectional type interoperability in the Cheerp compiler (leaningtech.com)

a year ago | ashvardanian | leaningtech.com | newest

6

NumPy vs. BLAS: Losing 90% of Throughput (ashvardanian.com)

a year ago | ashvardanian | ashvardanian.com | newest

11

Max Developer Edition Preview (modular.com)

a year ago | ashvardanian | modular.com | frontpage

1

CLoVe: Encoding Compositional Language in Contrastive Vision-Language Models (arxiv.org)

a year ago | ashvardanian | arxiv.org | newest

1

Intel Announces New Edge Platform for Scaling AI Applications (intc.com)

a year ago | ashvardanian | intc.com | newest

1

Lightbug: Simple and fast HTTP framework for Mojo (github.com/saviorand)

a year ago | ashvardanian | github.com | newest

2

USearch SQLite Extensions for Vector and Text Search (github.com/unum-cloud)

a year ago | ashvardanian | github.com | newest

0

VideoPrism: A Foundational Visual Encoder for Video Understanding (arxiv.org)

a year ago | ashvardanian | arxiv.org | newest

35

Measuring energy usage: regular code vs. SIMD code (lemire.me)

a year ago | ashvardanian | lemire.me | best

2

Borrow Checker, Lifetimes and Destructor Arguments in C++ (a10nw01f.github.io)

a year ago | ashvardanian | github.io | newest

1

Python Is Portable (2021) (ahgamut.github.io)

a year ago | ashvardanian | github.io | newest

2

Galileo v0.1 cross-platform map rendering engine released (maximkaaa.github.io)

a year ago | ashvardanian | github.io | frontpage

0

C++ Package Managers: The Ultimate Roundup (moderncppdevops.com)

a year ago | ashvardanian | moderncppdevops.com | newest

2

C++ Safety, Revisited (accu.org)

a year ago | ashvardanian | accu.org | newest

1

Development-Cycle in Cargo: 1.77 (rust-lang.org)

a year ago | ashvardanian | rust-lang.org | newest

6

What's Wrong with C++ Strings? (ashvardanian.com)

a year ago | ashvardanian | ashvardanian.com | newest

7

Show HN: StringZilla v3 with C++, Rust, and Swift bindings, and AVX-512 and NEON (github.com/ashvardanian)

a year ago | ashvardanian | github.com | newest