A developer's ongoing 32-week project to understand AI inference reached Phase 2, focusing on tensors, the core data structure of machine learning models. The developer demonstrated creating tensors i…
Perfetto's next release will improve PyTorch profile rendering by fixing a long-standing issue where overlapping events were dropped or poorly displayed. The previous workaround merged overlapping eve…
SambaNova Systems is hiring a Principal AI Systems Performance Engineer in San Jose, California, to optimize and scale state-of-the-art foundation models on its reconfigurable dataflow platform. The r…
AIsbom v1.3.0, a tool for scanning ML model artifacts, now flags Keras Lambda layers and embedded code objects as CRITICAL, without unmarshalling payloads. The tool also detects ONNX external-data pat…
A new open-source Rust project, tts-rs, delivers local text-to-speech on Apple silicon with three voice-cloning engines ported from PyTorch, achieving real-time factors as low as 0.252 for the Qwen3TT…
A new open-source PyTorch optimizer, the Riemannian Thermodynamic Optimizer, outperforms Adam on non-linear benchmarks by treating loss landscapes as physical topologies with Landauer dissipation and …
ReClaim, a computer-vision-powered lost-and-found system for campus communities, uses CLIP embeddings and cosine similarity to match lost items with found items. The system, built with FastAPI and PyT…
The RIS-Kernel repository implements the Reduced Interaction Sampling (RIS) framework, a model-agnostic runtime attention patching layer that enables running 64k+ token context windows on commodity CP…
Triton 3.7's new plugin extension system lets GPU kernel developers load custom MLIR compiler passes at runtime as shared libraries, eliminating the need to maintain a Triton fork. Meta's TLX extensio…
AMD's Ryzen AI Halo may outperform NVIDIA's DGX Spark for local AI development, according to a hardware comparison. The Ryzen AI Halo offers better power efficiency and thermal performance, while the …
A solo developer with no professional GPU background placed 12th of 183 in GPU MODE's batched QR decomposition contest, beating the cuSolver-backed baseline by 232x (419,000 µs down to 1,805 µs on NVI…
Sankalp placed 12th among 183 entrants with a time of 1,804.779 microseconds on an NVIDIA B200 in GPU Mode's qr_v2 competition, using OpenAI Codex to make over 1,500 submissions over 14 days. His resu…
A technical tutorial demonstrates an end-to-end supervised fine-tuning pipeline for the XYZ-Aquila-SFT dataset using Hugging Face Transformers, PyTorch, and PEFT, culminating in LoRA fine-tuning of Qw…
Divyakush Punjabi, a full-stack and AI engineer, explains self-attention, the core mechanism behind large language models and modern recommenders, without heavy math. He illustrates the concept with t…
Photoroom, a Y Combinator 2020 cohort company founded in 2019, is hiring a Product ML Engineer in Paris with a salary of €10k/yr, offering a hybrid work model. The role involves owning ML projects end…
A developer detailed how to build and run vLLM's Rust frontend on an AWS EC2 G5g instance with Graviton2 and an NVIDIA T4G GPU. The tutorial highlights that vLLM now requires a Rust toolchain for sour…
Convolution, a mathematical operation that slides a filter across data to extract patterns, underpins modern AI applications from Spotify's audio smoothing to autonomous vehicle pedestrian detection, …
WinCore 0.6.3, a free and open-source Python library for AI and PyTorch on Windows, has been publicly released, offering utilities for CPU/GPU management, memory handling, diagnostics, system detectio…
Researchers introduced Prof-K, a probabilistic one-pass filtering algorithm for top-k selection that guarantees correctness with probability at least 1 - ε, achieving 1.5x-10x speedups over PyTorch to…
PyTorch's torch._assert_async API enables asynchronous assertion on GPU device streams, avoiding graph breaks in torch.compile. The API accepts a boolean tensor and reports failures only when the GPU …