search
cod
Trends
- 1Dermatologist unveils 3D biophysical skin model built with AI coding●Show HN: I'm a dermatologist and I vibe coded a 3D biophysical skin model
A dermatologist has launched an interactive 3D biophysical model of human skin, saying he built it himself using AI-assisted 'vibe coding' despite having no formal programming background. The project, shared with the developer community, is drawing attention as another example of medical professionals using AI tools to build specialized educational and clinical resources without hiring engineers.
- 2
A Cape Cod Chronicle nature column describes a visit to the beach just before a storm arrives, reflecting on the shoreline's shifting mood as weather closes in. The piece is part of the paper's ongoing Nature Connection series exploring the local landscape. It is drawing attention among readers following coastal weather in the region.
- 3Cooking traditional fish dishes for mother in hospital●I've got some cod, ginger & onions slowly simmering in clay pot. Making Mom more traditional fish dishes to bring to the
A person is slowly simmering cod with ginger and onions in a clay pot, preparing traditional fish dishes to bring to their mother in hospital, believing fish is good for her recovery. They also brought her dragonfruit, papaya and roasted unsalted nuts and seeds the evening before, sharing the quiet acts of care publicly.
- 4Open-source model router aims at coding-agent performance●Show HN: Open-source model routing for coding agents at Astra-level performance
A developer has released an open-source model routing tool designed to send coding-agent requests to the best available models, claiming performance on par with Astra-level systems. The launch is being shared on Hacker News, where the community is debating whether the performance claims hold up and how practical the routing approach is for real-world coding workloads.
- 5Solar Phasing Offers 30% Off for Bandcamp Friday●We had a quantum propulsion issue during the trip for Bandcamp Friday. But.. until Monday, use the code "quantum" to get
The ambient music project Solar Phasing is running a promotion for Bandcamp Friday, offering 30% off everything in its catalogue until Monday with the discount code "quantum". The artist jokes about a "quantum propulsion issue" during the trip and thanks listeners for their support, inviting fans back on board for new sonic adventures.
- 6Stabbing reported on Flydubai flight from Dubai to Tel Aviv●DECODE : Dubai से Tel Aviv जा रही Flight में 'चाकूबाज़ी' | Sudhir Chaudhary | Flydubai | Israel
News anchor Sudhir Chaudhary's 'Decode' programme examined a knife attack aboard a Flydubai flight travelling from Dubai to Tel Aviv. The incident has drawn wide attention in India and the region, coming against the backdrop of heightened tensions around Israel and Gulf air routes, with viewers debating security screening and the motive behind the attack.
- 7
The t3code repository from the pingdotgg organisation is trending on GitHub, ranking among the platform's most-viewed repositories globally in the TypeScript category. Little detail is available beyond the project name, but its surge in traffic suggests the developer community is taking note of a new open-source tool from the team behind the T3 stack, created by developer Theo Browne.
- 8Developers debate whether AI coding agents produce good code▼Ask HN: Is anybody producing good code with coding agents?
A question circulating among developers asks whether anyone is genuinely producing good code with AI coding agents. The discussion reflects ongoing skepticism in the programming community about whether tools that generate code automatically can match the quality, reliability and maintainability of work written by experienced engineers, and whether real teams are getting production-grade results from them.
- 9
Sazabi has removed roughly 800,000 lines of unit tests from its codebase as part of a shift toward AI-assisted coding. The move has drawn attention among developers, with many debating whether large-scale test deletion is a sensible response to AI code generation or a risky erosion of software quality safeguards. Reactions are split between views that AI can replace traditional test coverage and warnings that regression bugs may go undetected.
- 10
A developer has published an account of spending a month using GLM 5.3 Flash, the fast model from China's Zhipu AI, as a daily coding assistant. The write-up covers what worked, what did not, and how the model compares with other assistants, prompting discussion among developers weighing cheaper open models against established options.
- 11Vibe-coded website that looks professionally designed●How our vibe coded website looks like a designer made it
A developer has written about building a website using 'vibe coding' — writing software by prompting AI rather than hand-crafting code — and argues the result looks as polished as if a professional designer had made it. The write-up is drawing attention among developers debating whether AI-assisted, prompt-driven development can now produce genuinely well-designed front-end work.
- 12Rhun: an open-source code editor written entirely in assembly●Show HN: Rhun, an open-source code editor written in assembly
A developer who goes by vladcodes has released Rhun, an open-source code editor written in assembly language. The project, hosted at rhun.app, is drawing attention on Hacker News, where commenters are debating why anyone would build an editor in assembly today, the performance implications, and the practicality of maintaining such low-level code.
- 13New CLI tool compresses tokens to cut AI coding costs●Show HN: Token compression CLI to save Codex/Astra costs
A developer has released a command-line tool that compresses prompts before they are sent to AI coding assistants such as Codex and Astra, aiming to reduce token usage and lower API bills. The launch is drawing attention on Hacker News, where users are debating how much compression can realistically save without degrading the model's output quality.
- 14New Agent Development Environment ships for coding agents●Agent Development Environment (ADE)and orchestrator shipping with coding agents
A developer tool called Cezar is shipping an Agent Development Environment (ADE) together with an orchestrator designed to work alongside coding agents. The launch is drawing attention among developers who follow agentic coding workflows, with discussion centred on how a dedicated environment and orchestration layer could improve how coding agents are run and coordinated on real tasks.
- 15
Claude Code, an agentic coding tool from Anthropic, is drawing attention on GitHub. The TypeScript-based tool runs in the terminal, understands a developer's codebase, and executes routine tasks, explains complex code, and handles git workflows through natural language commands. Developers are discussing it as part of the growing wave of AI-assisted coding assistants aimed at speeding up everyday programming work.
- 16Crocodile kills 11-year-old child in village attack▼⚠️Crocodile takes 11-year-old child😭Crocodile Attack Village Humans | Wildlife Attack Village Rescue
An 11-year-old child was reportedly taken by a crocodile in a village attack, with footage described as showing villagers responding to the wildlife incident. Such attacks typically occur in regions where communities live close to crocodile habitats, and rescue or recovery efforts often follow. The news is spreading widely as viewers share the distressing footage of the incident.
- 17Meta open sources code for DIY Muse AI gadgets●Meta open sources code to let you make Muse AI gadgets
Meta has released the code for Muse as open source, allowing anyone to build their own Muse AI gadgets. According to The Verge, the move lets developers and hobbyists access the underlying software for the AI-powered device rather than keeping it proprietary. The decision could spark a wave of third-party hardware experiments built on Meta's technology.
- 18Perspica launches as a semantic diff tool for code review●Show HN: Perspica – A semantic diff for reviewing code
A developer has released Perspica, an open-source tool that produces semantic diffs of code, aimed at making code review easier by highlighting meaning-level changes rather than raw line differences. The project is available on GitHub, and early discussion online centers on how well it compares to conventional diff and review workflows.
- 19C++ Insights tool shows how compilers transform source code●C++ Insights – See your source code with the eyes of a Compiler
C++ Insights, an open-source tool by Andreas Fertig, transforms C++ source code to show the transformations a compiler applies under the hood, such as template instantiations and implicit conversions. Developers on Hacker News are sharing the tool as a useful way to understand what the language actually generates, with its GitHub repository drawing renewed attention.
- 20Offrun launches to manage coding agents from one workspace●Show HN: Offrun – manage every coding agent from one workspace
A developer has launched Offrun, a new tool that lets users manage every coding agent from a single workspace. The project was shared on Hacker News as a Show HN post, where early commenters are taking a first look at the product and discussing its approach to orchestrating multiple AI coding assistants.
- 21Cameras catch dwarf crocodiles hunting bats in Gabon's caves●🐊 NEW: In # Gabon ’s # Abanda Caves, dwarf # crocodiles roughly 1.5 meters long live in permanent darkness among mounds
Remote cameras in Gabon's Abanda Caves have recorded dwarf crocodiles, about 1.5 meters long, catching bats mid-flight for the first time. The reptiles live in permanent darkness among mounds of bat guano, and while researchers already knew bats were part of their diet, no one had ever directly observed the hunt before.
- 22Graphene launches as data analysis toolkit for coding agents●Show HN: Graphene – Data analysis toolkit for your coding agent
A new open-source tool called Graphene has been introduced on Hacker News, described as a data analysis toolkit designed for coding agents such as AI programming assistants. The project is available on GitHub. Early engagement is modest, with a handful of upvotes as developers evaluate whether it fills a real gap in how agents handle data work.
- 23
A developer project called CodeGraph by colbymchenry is gaining attention on GitHub. It builds a pre-indexed knowledge graph of a codebase that automatically syncs as code changes, and works with popular AI coding assistants including Claude Code, Codex, Gemini, Cursor, Copilot and others. The tool runs fully locally and aims to cut token usage and reduce the number of tool calls needed when AI agents navigate large projects.
- 24Developer launches fast syntax-aware code diffing tool●Show HN: CodeDiff – Fast (<100ms), robust (99.95%) syntax-aware code diff
A developer has released CodeDiff, an open-source tool on GitHub that computes syntax-aware code diffs in under 100 milliseconds with a stated 99.95 percent reliability rate. It was shared on Hacker News as a Show HN post, where new tools and projects are presented to the community for feedback and discussion.
- 25
The US Coast Guard is searching for an air ambulance that went missing off Nantucket, Massachusetts. Details about the aircraft, the crew aboard, and the circumstances of its disappearance have not yet been confirmed, and the search operation is ongoing as rescuers try to locate any survivors in the waters south of Cape Cod.
- 26v0.7.0 Adds Generated-Code Admission With Explicit Verification Boundaries●v0.7.0 adds Generated-Code Admission, keeps verification boundaries explicit, and refuses to call unverifiable generated
Version 0.7.0 of an open-source software project introduces a Generated-Code Admission feature that keeps verification boundaries explicit and declines to label unverifiable generated code as safe. The release reflects a shift away from blindly trusting AI-generated code, requiring it to pass verification before being accepted. Early reactions from the developer community are positive, with discussion around testing, Python tooling, and how to handle machine-written code responsibly.
- 27Anthropic says AGI could arrive within a couple of years●AGI in a couple of years — Anthropic's RL leads on coding agents, Frontier Math and biology
Anthropic is claiming that artificial general intelligence could arrive within a couple of years, citing progress from its reinforcement learning systems on coding agents, the Frontier Math benchmark and biology tasks. The claim positions the company among the leaders in the race toward AGI and is drawing attention to how quickly frontier AI capabilities are advancing.
- 28
A rare September nor'easter struck Cape Cod, with officials highlighting that the storm's timing mattered for coastal impacts and emergency preparations. Local authorities on Cape Cod are drawing attention to how the early-season timing shaped the storm's effects on the region, an unusual occurrence for so early in the fall.
- 29Fiserv and FiCare Federal Credit Union clash over card-freeze passcodes●Fiserv, FiCare FCU clash over passcodes in card-freeze fight Fiserv's own lawyers say the company's system can unfreeze
Fiserv and FiCare Federal Credit Union are in a dispute over passcode requirements for freezing compromised cards. According to American Banker, Fiserv's own lawyers acknowledged in the case that the company's system can unfreeze a card flagged for fraud before the call reaches an agent. The fight centers on who controls card-freeze safeguards and what protections customers actually get when fraud is suspected.
- 30Developers Delete Unit Tests as AI Coding Agents Spread●Developers Delete Unit Tests as AI Agents Reshape Coding
Software developers are reportedly removing unit tests from their codebases as AI coding agents take on more of the programming work. The practice is drawing debate, with some arguing tests become unnecessary when AI generates and verifies code, while others warn deleting tests undermines safety nets and long-term code quality.
- 31Developers Delete 800,000 Lines of Unit Tests in AI Shift●Developers Delete 800,000 Lines of Unit Tests for AI Coding Shift
Software developers have removed roughly 800,000 lines of unit tests from their codebase, citing a shift toward AI-assisted coding practices. The move has sparked debate among engineers about whether traditional test suites still add value when AI tools generate and validate code, with critics warning it could undermine software reliability and long-term maintainability.
- 32
The classic striped UPC barcode may soon disappear from retail shelves. A piece in The Hustle argues that traditional one-dimensional barcodes are nearing the end of their fifty-year run, as retailers and standards bodies push toward richer 2D codes that can hold far more product information. Discussion is focused on what the transition means for shoppers, stores and the global supply chain.
- 33Meta opens up code so users can build their own AI gadgets●Meta wants you to build your own AI gadget, and it’s giving away the code
Meta is releasing code that lets people build their own AI-powered gadgets, according to Gulf News. The move signals the company's push to get its AI software into more devices by letting developers and hobbyists create custom hardware rather than relying on Meta's own products, opening the door to a wider ecosystem of third-party AI devices.
- 34
Software developers are increasingly removing unit tests from their codebases, arguing that AI coding assistants make them unnecessary. Supporters say AI can regenerate and verify code on demand, while critics warn that deleting tests removes safety nets, makes regressions harder to catch, and signals a worrying shift in engineering practice.
- 35Working Code Isn't Understanding: Educators Question What Kids Learn From Coding●A child finishes a coding project, presses run, and everything works. That looks like success. Then someone asks, “Can y
A child completes a coding project and it runs perfectly, but struggles when asked to explain what the code actually does. Educators argue this gap between producing working programs and genuine comprehension is a real problem in computer science education. Completing a project, they say, does not guarantee a student understands the concepts behind it, and explanation should be part of how learning is assessed.
- 36Meta open-sources Muse tools so users can load AI agents onto their own devices●Meta has opened up the code of its tools for Muse. Now you can "populate" the AI agent with your own gadget
Meta has released the code of its tools for Muse, allowing users to run the AI agent on their own gadgets. The move gives developers access to the underlying tooling, letting them 'populate' the Muse agent directly on personal hardware rather than relying on Meta's own infrastructure.
- 37Codex plugins can now be used inside Pi coding agent●Show HN: Use all Codex Plugins inside Pi I just realized that codex now exposes local server endpoints for all plugins w
A developer has discovered that Codex exposes local server endpoints for all of its plugins without extra authentication, meaning those plugins can be used from any other model or agent harness. A new Pi install package lets users connect to all Codex plugins with a single auth setup. Developer communities are discussing what this means for interoperability between AI coding tools and whether open local endpoints could raise security questions.
- 38Altucher says electricity, not chips, is AI's real limit▼James Altucher: The Real Ceiling on Artificial Intelligence Isn't Chips or Code. It's Electricity, and He Says Elon Musk Just Found a Way Around It.
Entrepreneur and author James Altucher argues that the true bottleneck for artificial intelligence is not semiconductor supply or software, but electricity generation. In a widely syndicated column, he claims Elon Musk has found a way around this power constraint, a view feeding ongoing debate about whether energy infrastructure, rather than chip availability, will decide how fast AI can scale.
- 39Meta Releases Open Code for Building Muse AI Devices▼Meta Unveils Open Code for Creating Muse AI Devices
Meta has made open code available for creating Muse AI devices, according to a report by Nacionale News. The move suggests the company wants outside developers and hardware makers to build on its Muse technology rather than keeping it proprietary. Details on licensing, supported hardware, or launch timelines were not included in the initial coverage.
- 40
Development teams are removing around 800,000 lines of unit tests from their codebases, a move tied to the shift toward AI-assisted coding. The story has sparked debate among engineers: some argue AI-generated code changes how testing should be structured, while others warn that deleting tests undermines software safety and maintainability. Commenters are divided on whether it reflects progress or a risky shortcut.
Repos
- pingdotgg/t3code
- colbymchenry/codegraph Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGrav
- anthropics/claude-code Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster
- microsoft/vscode Visual Studio Code
- zai-org/ZCode Z.ai's coding agent harness. Powerful, intelligent, extensible.
- anthropics/claude-code-action
- wileai/pi-codex-connectors Use every connected Codex (ChatGPT) connector from any pi model
- TianyuCodings/NanoJev A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline.
- egma-ai/jev-code-reviewer Review behavior, not just diffs. Jev prioritizes human attention; OpenAI explains the changes. Local CLI + agent skill +
- gylive/ccodex-sleep-state 尝试改善 Codex 降智、限流与连接体验。本地一键启动、网页配置修复,支持 Astra / Sol / Terra、订阅和代理;不保证取得指定 state 或提升模型质量。
- ivankovic/codediff Fast, robust, accurate diffing