PersonWise

The deck was the easy part — someone still has to present it.

Kairo

Constrain LLM inference paths using measured RTX 5090 results and correctness checks.

An evidence-based research workbench that helps Blackwell inference researchers promote only results measured and verified for specific models, workloads, and environments into runtime policy.

Neopress | AI Website Builder for SEO, GEO & Growth

Create pages through conversation, publish structured content, and improve your site with SEO, GEO, forms, and analytics in one workflow.

Image25

Generate and edit images online with GPT Image 2.5.

study-kit

Focused learning with Claude Code and Codex.

Sentralis

Professional scenario, risk, liquidity, and counterparty analytics for self-directed crypto holders, with a risk workbench for deep investigations, scheduled reports and…

getdeck

A web tool for designing AI presentations with varied slide layouts and themes

A web design application for users who want to build presentations quickly by previewing 173 slide layouts and 41 themes, editing in real time and exporting the result.

agent-dev-team

A development team setup for AI coding tools with role levels and handoff procedures.

For development teams defining task responsibilities in AI coding tools, it provides graded agent roles, reusable workflow skills, and procedures for handing work to a higher-level role.

Hacker News, Without AI

A simple Hacker News feed that filters out AI-related content

For readers who want to see Hacker News posts excluding AI content, it provides submissions that pass the filter and an RSS subscription option.

learnlance

Learn from your AI coding agents.

Kibble

Read short AI updates by topic, with source links

A reading service for people who want a quick overview of changes in AI. It provides short posts about AI news, learning resources, code, and models, each with a source link.

Vehla

One app instead of fourteen.

termique

An SSH client that syncs all your devices without exposing credentials in plain text

Developers managing multiple servers can use a terminal, SFTP, key management and AI assistance in one place, and sync hosts and credentials between desktop and mobile devices with end-to-end encryption.

no_human

A local development tool that turns tickets into pull requests through planning and verification.

A free, open-source tool for development teams turning issue tickets into code changes. It creates acceptance criteria before coding and prepares pull requests after a separate model's review and test results.

Sealed Rose

A detection tool that checks videos frame by frame for deepfakes and traces of AI generation

Users checking a video's authenticity can upload a clip or paste a direct link to analyze face swaps and traces of AI synthesis, and see a likely generation source.

sorify

Sorify - AI Agentic QA Platform for AI agents.

graphify-csharp

An agent tool that finds call and reference relationships in C# code using compiler analysis.

For coding agents analyzing C# codebases, it reads projects with Roslyn and MSBuild and extracts precise caller, reference, implementation and inheritance relationships.

Codeberg.org

A technical book exploring the reasons for and understanding of programming in the age of AI.

An English-language technical book for readers who want to examine programming as a way of thinking about language, knowledge and creativity, beyond the craft of making software.

DiskTriage

A CLI for safely clearing unnecessary files from the Windows C drive

Helps Windows users whose system drives are running low because of development tools, AI models, package caches or game stores find large items to delete or move to another drive.

hcker.news

A Hacker News reader for searching stories and comments while filtering out noise

A reader for people following Hacker News stories and discussions who want to search stories and comments and filter out material they are not interested in.

TokenDelivery.ai

A model service that provides byte-for-byte identical answers to the same request.

It says it runs open-weight models deterministically to deliver the same answer every time for users who need consistent results from repeated requests.