
JustHireMe Ranks and Tailors Job Applications Entirely on Your Device
A desktop AI workbench keeps your resume, leads, and generated documents on your machine while ranking job fit with explainable graph proofs.
20 stories

A desktop AI workbench keeps your resume, leads, and generated documents on your machine while ranking job fit with explainable graph proofs.

Google's open-source inference stack pushes Gemma 4 to 76 tokens/sec on-device, adds Swift and JavaScript APIs, and powers Chrome and Pixel Watch

Codeg pulls Claude Code, Codex, Gemini CLI, and half a dozen other coding agents into one workspace with delegation and shared history.

An open-source macOS dictation app runs Whisper and Parakeet locally, offering a one-time-purchase alternative to Wispr Flow and Superwhisper.

OS1 Edition wraps Nous Research's Hermes Agent into a Her-inspired macOS app with one-click cloud VMs and WebRTC voice control

A self-hosted, all-in-one AI companion bundling 3D VTuber avatars, computer-use agents, IM/livestream bots, and an OpenAI-compatible local API.

LeanCTX is a single Rust binary that sits between AI coding agents and your files, cutting token usage 60-90% via smart compression and caching.

A lightweight Python MCP server gives AI agents free DuckDuckGo search and webpage fetching, with a new curl backend to slip past Cloudflare bot filters.

A Rust CLI collapses git worktree management into three commands, letting you run Claude Code and Codex agents side by side without stepping on each other.

A self-hosted open-source platform that normalizes health data from Garmin, Whoop, Apple Health, Oura, and more behind a single REST API.

The open-source Go bridge lets you drive Claude Code, Codex, Cursor and Gemini CLI from Feishu, Telegram, Slack, Discord and nine other chat apps.

HiDream open-sourced an 8B pixel-native transformer that unifies text, images, and task conditions in one shared token space at up to 2K resolution.

Repomix bundles an entire repository into a single AI-friendly file, with Tree-sitter compression, MCP server support, and security scanning built in.

OpenBMB's 1.3B multimodal model beats Qwen3.5-0.8B by 3 points on Intelligence Index while using 19x fewer tokens and runs on phones.

HumanSignal open-sourced Adala, a Python framework where agents iteratively learn data labeling skills from ground truth datasets using LLM runtimes.

An open source terminal agent runs paper search, multi-agent deep research, replication audits, and citation verification from a single CLI.

Google DeepMind's curated library of physics-ready robot models now spans 60+ humanoids, quadrupeds, arms and grippers with GPU-friendly MJX variants.
Nous Research's Lighthouse Attention cuts long-context LLM pre-training time by 1.4–1.7x, running 17x faster than standard attention at 512K context — with no custom sparse kernel and no quality loss at inference.

Olaf Hohm releases master-level lecture notes tracing the shared mathematical language of spin glasses, Boltzmann machines, and modern LLMs

OpenSwarm ships an open-source, terminal-native team of eight coordinated agents that turn one prompt into decks, reports, videos, and images.