18d ago
$5k in compute and LLM credits on JackHamr for the first 30 qualified startups
JackHamr platform offering $5k compute and LLM credits to qualified AI agent startups with infrastructure support.
18d ago
Concrete Problems in AI Safety – Dario Amodei (2016) [video]
2016 video presentation by Dario Amodei on concrete safety problems in AI systems.
18d ago
Small aircraft crashes into Beijing's tallest building, videos show
News article about aircraft crash unrelated to AI/tech.
18d ago
Why memory is not enough: you need context management
Technical article on context management vs memory in AI agents, addressing context window overflow challenges.
18d ago
Comparing web search API providers on a Deep Research gauntlet
Benchmark comparing web search API providers using LLM judges on 32 real-world agentic research tasks with ELO scoring.
18d ago
Literate Programming in the Age of LLMs
Explicode: tool enabling literate programming by embedding rich Markdown documentation in code comments for humans and AI agents.
18d ago
What if ideas aren't getting harder to find, after all?
Discussion prompt about research productivity and whether finding new ideas is becoming harder.
18d ago
Give GitHub Copilot CLI real code intelligence with language servers
Guide on integrating language servers with GitHub Copilot CLI for improved code intelligence.
18d ago
Show HN: Statey – the database your AI shares across every chat, over MCP
Statey database tool enabling AI agents to share persistent state across chats via Model Context Protocol.
18d ago
MirrorCode: What's the largest software project AI can complete on its own?
MirrorCode benchmark co-developed with METR testing AI models on long-horizon full-program reimplementation tasks without source code.
18d ago
Amazon Q flaw let booby-trapped Git repos execute code, swipe cloud creds
Security vulnerability in Amazon Q AI coding assistant allows executing code and stealing cloud credentials through malicious Git repos via Model Context Protocol mishandling.
18d ago
Hush, let an AI agent use your secrets without ever seeing them
Secret store for AI agents that prevents plaintext exposure in logs, transcripts, or cloud systems.
18d ago
Framesmith 1.7 – a quality gate that tells an AI agent when a UI is done
Open-source MCP server providing visual canvas for AI coding agents to design UI before framework implementation.
18d ago
Open AI SDR – An Open Source Version of the Artisans Ava BDR
Open-source AI sales development representative agent that sources leads, writes outreach, and books meetings.
18d ago
Cairn – self-hosted health OS that connects labs, training and meals
Self-hosted health tracking application connecting lab results, training, and nutrition data.
18d ago
Show HN: Mantis, A self-hosted LLM gateway
Self-hosted LLM gateway for small teams with single-command AWS deployment.
18d ago
VibePHP
Satire: VibePHP runtime where AI 'runs PHP in its head' instead of executing code, producing non-deterministic results.
18d ago
Building Voice AI Workflows with Branches Instead of One Giant Prompt
Tutorial building voice AI workflow for insurance claims intake using branching logic instead of monolithic prompts.
18d ago
Summary of METR's predeployment evaluation of GPT-5.6 Sol
METR's independent external evaluation of GPT-5.6 Sol on software task benchmarks. Conducted under NDA with OpenAI review approval.
18d ago
Akrites: The Latest Attempt to Protect Open-Source from AI Attacks Has Arrived
Linux Foundation initiative to identify and fix vulnerabilities in open-source software threatened by AI-accelerated attacks.
18d ago
A list of software and other offerings with free developer tiers
Curated list of software services offering free developer tiers for infrastructure developers (SaaS, PaaS, IaaS).
18d ago
My-Pi Coding-Agent
My-Pi is curated Pi coding-agent distribution with MCP, LSP, skills, recall, redaction, telemetry, and team mode preinstalled.
18d ago
GPT-5.6 Preview System Card
OpenAI's technical documentation for GPT-5.6 family (Sol, Terra, Luna). Details safety safeguards and planned general availability timeline.
18d ago
U.S. government will decide who gets to use latest upgrade to ChatGPT
Brief headline about U.S. government ChatGPT access restrictions without substantive content.
18d ago
Murmur: Shared communication bus for your coding agents
Murmur is shared communication bus for coding agents. Multiple LLM agents (Claude, Copilot, etc.) communicate via single MCP HTTP daemon with @-mentions.
18d ago
Ventora Expands Its AI Business Builder to Help Solo Founders
Business platform helping solo founders with non-technical aspects beyond AI code generation like validation and positioning.
18d ago
Bigger context windows are the wrong abstraction for coding agents
Analysis arguing coding agents need durable codebase memory over larger context windows for effective development.
18d ago
Hasp – Local Secret Broker
Hasp is a local secret broker for AI agents/apps. Encrypts secrets in vault, controls access by project boundary and time window, never exposes raw values.
18d ago
The Discoverable Evidence of AI-Assisted Software Porting
Case study of AI-assisted software porting from one language to another with code generation artifacts.
18d ago
Please don't use an LLM to communicate with other human beings
Opinion piece arguing against using LLMs for human communication, emphasizing importance of personal writing skills.
18d ago
GPT-5.6 System Card [pdf]
Link to GPT-5.6 System Card PDF with minimal description.
18d ago
Ask HN: How do you know when your AI agent's output quality degrades?
Ask HN discussion on monitoring AI agent output quality degradation in production beyond cost/latency metrics.
18d ago
Ask HN: Options for an independent AI researcher with strong results?
Ask HN thread from independent AI researcher with novel analytical framework for neural representational models seeking career options.
18d ago
Hallucinate.md: The open standard for telling AI not to hallucinate
Open standard markdown file (hallucinate.md) to instruct AI coding agents not to hallucinate. Simple specification for repository integration.
18d ago
OpenAI releases GPT-5.6 to select users vetted by US Government
News headline about OpenAI releasing GPT-5.6 to government-vetted users. Limited information provided.
18d ago
Previewing GPT‑5.6 Sol: a next-generation model
OpenAI announces GPT-5.6 series with three models: Sol (flagship), Terra (balanced, 2x cheaper), Luna (fast, affordable). Includes safety improvements.
18d ago
AI coding agents could soon cost more than the developers using them
Gartner analysis of escalating costs from AI coding agents shifting to consumption-based pricing, creating unpredictable developer expenses.
18d ago
AI Learns the "Dark Art" of RFIC Design
AI used to design RF integrated circuits faster than traditional methods. Focus on hardware design, not core AI/ML interests.
18d ago
Show HN: TBD, a Mac-native CLI-forward coding agent multiplexer
Mac-native CLI-forward coding agent multiplexer with GUI, designed to expose all functionality for automation and multi-agent coordination.
18d ago
White House Will Ad Hoc Decide Who Can Individually Access GPT-5.6
Opinion piece criticizing White House ad hoc decision-making process for frontier AI model access as lacking transparency and formal procedures.
18d ago
Show HN: Vynex API – One endpoint for 34 LLM models, paid with USDT
Unified LLM gateway supporting 34 models via single OpenAI-compatible endpoint with per-token pricing and USDT payment.
18d ago
Combining LLMs Rarely Beats the Best Single Model, I tested 67 frontier models
Research comparing 67 frontier LLM models showing combining models rarely outperforms best single model.
18d ago
Show HN: Smart model routing directly in Claude, Codex and Cursor
Model router that intelligently routes requests to optimal LLM for coding agents like Claude, Codex, Cursor to reduce costs.
18d ago
A free checker for whether AI search engines can cite your site
Free tool checking if websites are discoverable and citable by AI search engines like ChatGPT, Claude, and Perplexity.
18d ago
Show HN: AI Is Too Good at Coding, Let's Make It Better
persist-os CLI tool that records AI coding decisions and architecture to enable better code quality and scalability beyond rapid prototyping.
18d ago
OpenAI set to limit GPT 5.6 rollout after call from Trump administration
News article about OpenAI GPT model rollout policy.
18d ago
The Coming Divide: AI-Native or Left Behind
Opinion piece about AI hype cycle and concerns regarding venture capital investment in AI startups versus dismissing it as another bubble.
18d ago
Getting AI credits for reading ads while your coding agent is doing its thing
Concept about earning AI credits while coding agents work. Title only, minimal detail.
18d ago
Prompts in Manuscripts Exploit AI-Assisted Peer Review
Research title only about AI-assisted peer review vulnerabilities. Insufficient content for evaluation.
18d ago
Show HN: `uvx ptn` and expose any system to agents (dangerously)
Tool enabling agents remote access to Linux/Mac/PC systems via simple CLI with phone-based kill switch.