All Articles
17533 articles total
OpenAI no longer recommends SWE-Bench Pro
OpenAI has shifted its recommendation away from the SWE-Bench Pro benchmark for evaluating software engineering capabilities of AI models.
Cloudflare Drop
Introduction to Cloudflare Drop, a tool for managing and deploying assets and configurations via Cloudflare's edge network.
FAANG Simulator
A simulation tool designed to mimic the experience and environment of working at FAANG companies.
Show HN: Onboard-CLI, a LLM powered and AST-based tool to visualize codebase
Onboard-CLI is an LLM-powered, AST-based command-line tool used to visualize the structure and logic of a codebase.
Turning a pile of documents into a searchable useable knowledge base
A guide or discussion on the process of converting unstructured documents into a searchable and functional knowledge base.
Show HN: Hnwork.app – UI for Who is hiring posts
Hnwork.app is a community-driven user interface for navigating and filtering 'Who is hiring' posts from Hacker News.
Agentic test processes, LLM benchmarks, and other notes on agentic coding fr
Exploration of agentic test processes and LLM benchmarks specifically focused on the evolution of agentic coding frameworks.
Sometimes it is the network: a war story
A technical 'war story' detailing a debugging experience where the root cause of a system failure was identified as a network issue.
Google’s deepfake detector system used to debunk McConnell hoax pic
Google's deepfake detector system successfully identified and debunked an AI-generated fake image of Senator Mitch McConnell.
A bug which affected only left handed users
A discussion about a software bug that specifically affected only left-handed users.
I Built a Telegram Client for Pi
A developer shares their project of building a custom Telegram client specifically for the Raspberry Pi.
Show HN: Agent Draw: An agent draws while you talk, built on TLDraw
Agent Draw is an AI-powered agent that draws in real-time based on voice input, built on top of TLDraw.
SpaceXAI releases Grok 4.5, which Elon describes as an ‘Opus-class model’
SpaceXAI has released Grok 4.5, which is described as a more efficient and cheaper 'Opus-class' AI model.
‘Slow-cial’ app Roost forces you to slow down to the speed of a carrier pigeon
Roost is a 'slow-social' app designed to mimic the speed of carrier pigeons to combat fast-paced online culture.
This startup thinks robotics is about to have its ChatGPT moment
General Intuition is leveraging massive amounts of video game data to train foundation models for physical AI and robotics.
When Should LLMs Search? Counterfactual Supervision for Search Routing
Researchers propose a method for search routing in LLMs using counterfactual supervision to decide when external search is actually beneficial.
Data-dependent Evaluations for Budgeted Submodular Maximization
This paper introduces new data-dependent upper bounds for budgeted submodular maximization to better evaluate algorithmic performance.
LEGATO 2: Toward Multimodal Sheet Music Recognition and Understanding
Legato 2 is a multimodal pipeline for optical music recognition that extracts symbolic notation and semantic knowledge from sheet music images.
FORGE: Towards Functional Tool-Use Generalization via Keypoint Trajectory Reasoning
FORGE is a two-stage policy that enables robots to generalize functional tool use by reasoning through keypoint trajectories.
OpenMandriva: Statement regarding attempted distribution sabotage
OpenMandriva issued a statement regarding an attempted sabotage of their distribution.