#INTERfaces

Your AI agents now have their own little screen on your desk. Meet Harness Pro by Autonomous. A dedicated desk display that lets you monitor and control AI coding agents running across multiple computers. The interesting part is that it uses an ESP32-P4 rather than a conventional mini PC. It has a 3.95-inch, 720 × 720 touchscreen, three microphones and a speaker, all inside a compact CNC-machined aluminium enclosure. It connects over USB-C to a computer running the Harness app. You can switch between Claude Code, Codex, Cursor and other agents, see what they're doing and tap the screen to give them voice instructions. Even if you close the app or the device, your agents continue working on their machines. What I particularly like is the hardware design. The ESP32-P4 handles a dedicated interface for something that normally lives inside a developer's terminal. Building physical interfaces for AI agents is worth to check out for.

Harness Engineering 2.0 is here! After protocols for tools, agent communication, and agent interfaces, harnesses now have one too. Developers just built UHP (Unified Harness Protocol), a standardized interface for running agent tasks across any agent harness. HarnessRouter is its open-source server implementation. Here’s how it works: - The product backend sends a task with a `harness_id`. - HarnessRouter starts the selected harness in its own session. - The harness runs its native agent loop and streams back progress. - Files, artifacts, failures, and the final output return through the same interface. Without UHP, integrating every harness needs separate handling for task execution, sessions, streaming, files, recovery, and output formats. With UHP, Codex, Claude Code, Gemini CLI, Hermes, Pi, and other harnesses can plug into the same product backend while keeping their native tools, skills, models, and agent loops. This means developers can: - Switch harnesses without rebuilding the product integration. - Run the same task across different harnesses. - Add new harnesses without maintaining another execution stack. HarnessRouter currently supports 30 harness bases. Here’s the repo: (don’t forget to star it ⭐) As per official benchmarks, changing only the harness while keeping the model fixed reduced costs by 25% to 64% across several healthcare compliance tasks. And changing both the harness and model widened the difference to 99.8% between the highest and lowest-cost configurations. Developers can test their own workloads across different configurations using Harness Arena. Alibaba Cloud's Qoder has also adopted UHP, making it available as an agent backend through the HarnessRouter API. If you want to dive deeper, I already wrote an article that explains why model routing is different from harness routing and what supporting multiple agent runtimes actually involves. It also covers UHP, the local setup, a working API call, and how sessions and files work. Read it below.

CCNA 2.0 and what to expect: > Interface and cable troubleshooting (copper and fiber) > Hypervisors, virtual machines, and containers > IPv4 addressing, assignment, and subnetting > IPv6 addressing, assignment, and prefix sizing > Wireless principles, bands, channels, RF characteristics, security, and interference > Wired and wireless client connectivity troubleshooting > DHCPv4 client, server, and relay troubleshooting > Layer 2 and Layer 3 interfaces > 802.1Q trunking > LACP and EtherChannel > Switch virtual interfaces (SVIs) > Edge-host connectivity, VLANs, and PoE > Wireless access points and controller-based networking > Voice over IP (VoIP) phone connectivity > Virtualized hosts and network appliances > CDP and LLDP > Network troubleshooting with show commands, ping, traceroute, and packet captures > Rapid PVST+ > STP root bridge, root port, port roles, and port states > PortFast, root guard, loop guard, and BPDU guard > Routing table interpretation > IPv4 and IPv6 static routing > Default, network, host, and floating static routes > Single-area OSPFv2 for IPv4 > OSPFv3 for IPv6 > OSPF neighbor adjacencies, point-to-point, broadcast, DR/BDR, and router ID > First Hop Redundancy Protocols (HSRP and VRRP) > Local usernames and AAA with TACACS+ and RADIUS > SFTP and SCP > NAT and PAT on IOS XE > DNS record types (A, AAAA, CNAME, MX, NS, PTR) > IPsec remote-access and site-to-site VPNs > IPv4 ACLs (standard, extended, numbered, and named) > DHCP snooping > Dynamic ARP Inspection (DAI) > Storm control > IPv6 Router Advertisement Guard (RA Guard) > Port security > Agentic AI for network operations > Generative AI prompt engineering for network operations > Network management approaches (device-based, cloud-based, controller-based, automation-based, and Infrastructure as Code) > SNMP > Ansible configuration management > Syslog message content, severity levels, and facilities

The Antigravity interface has a new look & feel. Many of you have already seen it, so let us know what you think!

The “Trojan horse” may not look like a conventional chip anymore. It may look like a biosensor. A biosensor is designed to detect biological information and convert it into a measurable signal. Depending on the system, biosensors can be designed to monitor things like: Electrical activity. Chemical changes. Temperature. Pressure. Movement. Glucose. Hormones. Proteins. Neural activity. Muscle activity. Other physiological signals. Now take that basic concept and shrink it. Technology that once required visible electronic components has been getting smaller for decades. Today, researchers are developing microscale and nanoscale biosensors, injectable electronics, semiconductor nanowires, neural interfaces and wireless systems designed to interact directly with biological tissue. That is where the discussion gets much bigger. A biosensor doesn’t have to look like the traditional implant people imagine. It may be microscopic. It may be flexible. It may be distributed across multiple locations. It may be designed to sense biological activity. It may communicate with an external system. It may operate as part of a larger network. And depending on the architecture, it may be capable of collecting information from the body and transmitting that information elsewhere. UC Berkeley researchers developed “neural dust”, tiny, wireless, batteryless sensor systems designed to interface with nerves and muscles and communicate using ultrasound. Nature Nanotechnology published research demonstrating syringe-injectable mesh electronics capable of integrating with biological tissue and recording neural activity. DARPA’s N3 program pursued high-performance bidirectional brain-machine interfaces designed to both READ FROM and WRITE TO neural activity using optical, acoustic and electromagnetic approaches. And then there are Wireless Body Area Networks. IEEE 802.15.6-2026 covers short-range wireless communication in the vicinity of, or inside, the human body. Think about what that means. A biosensor is not just a tiny object. It can be part of an information system. A biosensor can detect something happening biologically. That signal can be converted into data. That data can potentially be transmitted to another device. Multiple sensors can potentially function together. A gateway can potentially move that information into a larger network. Now add AI, cloud computing, neural interfaces and advanced wireless communication. The question becomes much bigger than: “Is there a chip?” The better questions are: Are there biosensors? What are they sensing? What biological information can they collect? Can they communicate? Can multiple biosensors operate together as a network? Can they interact with the nervous system? Can the same architecture that reads biological activity also be paired with systems designed to stimulate or influence biological activity? Who receives the data? Who controls the network? And most importantly: Did the person consent? That's where the “Trojan horse” argument becomes serious. The concern isn’t that every biosensor is dangerous. Biosensors have legitimate medical, scientific and therapeutic applications. The concern is what happens if a technology capable of sensing, communicating with or interfacing with the human body is ever introduced, deployed or used without informed consent. That changes the entire conversation. Medicine becomes surveillance. Monitoring becomes data collection. A therapeutic interface can potentially become a control interface. And the person may not even know the system is there. When people hear “nanoscale biosensor” and dismiss it as science fiction, they’re ignoring an entire field of bioelectronics that already exists. Neural dust is real. Injectable electronics are real. Biosensors are real. Wireless Body Area Networks are real. Semiconductor nanowires are real. Bidirectional brain-machine interfaces are real. The question is what happens as these technologies converge and become smaller, more powerful and more connected. Explore the research, patents, government records and technical documents: THE GHOST STUDY This is one reason the published GHOST study deserves serious attention. The study discusses its findings alongside established research involving semiconductor nanowires, injectable mesh electronics, neural dust, biosensing, DARPA N3 and Wireless Body Area Networks. Twenty-two participants were scanned over 63 days using standardized EDD-24XT NLJD procedures. At baseline, all 22 showed second-harmonic-dominant responses classified by the instrument as “Probability Silicon.” After 63 days of MasterPeace use: Mean Signal Intensity Score decreased approximately 79%. Active Signal Spots decreased approximately 67%. Mean Signal Intensity per Detection Site decreased approximately 38%. READ THE PUBLISHED GHOST STUDY: Those measurable changes are one of the biggest reasons people following this research want to be on MasterPeace. Create your free MasterPeace account: So maybe we need to stop asking whether someone has a conventional “chip.” Maybe we should be asking whether modern biosensing technology has already moved far beyond what most people picture when they hear that word. How small can biosensors become? What can they measure? How can they communicate? How long can they function? Can they be networked? Can they interact with neural activity? Can they transmit biological data outside the body? And who is making sure this technology can never be used on human beings without informed consent? Get on the Mind Nexus email list: FULL INTERVIEW: PRIVATE ONE-ON-ONE SCANNING & DOCUMENTATION — RENO, NEVADA Mind Nexus offers private RF and NLJD Scanning & Documentation appointments for people seeking objective documentation of anomalous frequencies and nonlinear responses. Phase 1 — Scanning & Documentation RF scanning documents anomalous frequencies around a person and possible emissions within the surrounding environment. NLJD scanning uses nonlinear harmonic responses to look for semiconductor-type electronic anomalies, including electronic components that may not be actively transmitting. Phase 2 — Coaching & Next Steps may be available after results are reviewed and can include coaching, frequency research, controlled verification, expert review, additional technical analysis and identifying outside professionals. Scanning is NOT a medical diagnosis. Our scans detect anomalous frequencies around a person and within the environment, not inside the body. Findings are for informational purposes only and can support personal investigations. Petition · Stop Non-Consensual Human Testing Goal: 1,000,000 signatures: Facebook: @MindNexusLive Instagram: @MindNexusEvents X: @MindNexusLive YouTube: @MindNexusLive Rumble: @MindNexusLive #MindNexus #Biosensors #Nanosensors #NeuralDust #Bioelectronics #Nanotechnology #Neurotechnology #WirelessBodyAreaNetwork #BrainComputerInterface #GHOSTStudy #MasterPeace #NLJD #CognitiveLiberty #HumanAutonomy

@RealAlexJones handed the fourth hour of his show over to Mind Nexus, and all three founders, Jesse Beltran, Mike Grady, and Deserie Foley, used that hour to lay out what we’ve been documenting, why we believe AHI/Havana Syndrome-type experiences extend beyond government personnel, and what people can actually do next. @INFOWARS We started with the government’s own record. Congress held the 2024 hearing “Silent Weapons: Examining Foreign Anomalous Health Incidents Targeting Americans in the Homeland and Abroad,” with former Pentagon investigator Lt. Col. Greg Edgreen among the witnesses. The National Academies had already concluded that directed, pulsed RF energy was the most plausible mechanism among those it considered for some of the distinctive Havana Syndrome cases. That’s the foundation. Then we moved into what Mind Nexus has actually been finding. Jesse walked through the technical side of our work, RF detection, spectrum analysis, Non-Linear Junction Detection, the EDD-24XT, patents, scan results, case documentation, and examples involving material that was later physically removed and examined. We showed the equipment. We showed the reports. We showed patents. We showed congressional testimony. We showed research. We showed images and case material from people who came to us because they were experiencing things they couldn’t explain. The point wasn’t to tell people what to believe. The point was to ask what can actually be detected and documented? Government employees experiencing AHI have received congressional hearings, specialized medical attention, intelligence investigations and federal resources. GAO later interviewed 65 AHI patients about their experiences navigating treatment. But what happens when a civilian reports similar pressure, tones, vibrations, neurological disruption, unexplained RF activity or other unusual experiences? Who scans their environment? Who preserves their data? Who checks for repeatable technical anomalies? Who takes them seriously enough to investigate? Mike connected those findings to the congressional record, public research, government documentation and a basic question that still isn’t being answered... If AHI is serious enough for congressional hearings when government personnel report it, what happens when civilians report similar experiences? Deserie brought it back to the human side,families, people looking for answers, and the need for a real path forward instead of endless fear and confusion. And Jesse showed what that path begins with... Measurement. Documentation. Repeatability. Evidence. We also talked about solutions. That includes reducing unnecessary environmental and toxic burden and supporting the body’s natural elimination pathways. We discussed MasterPeace as one of the wellness tools people can explore as part of that broader approach. This broadcast was important because we weren’t talking around the issue. We put the research, scans, patents, government hearings, case material and questions on the table for MILLIONS OF PEOPLE to examine themselves. If Havana Syndrome deserves investigation when it happens to diplomats, intelligence officers and military personnel, why shouldn’t unexplained AHI-type reports from civilians be technically investigated too? What if the biggest mistake is assuming we already know the answer before anyone takes measurements? Watch. Research. Ask questions. Then follow the evidence. Get on our email list: One-on-one scanning: Our documentation process includes RF detection and Non-Linear Junction Detection using the EDD-24XT. We document anomalous frequencies around a person and possible emissions within the surrounding environment. Scanning is NOT a medical diagnosis. Our scans detect anomalous frequencies around a person and within the environment, not inside the body. Findings are for informational purposes only and can support personal investigations. Petition · Stop Non-Consensual Human Testing Goal: 1,000,000 signatures: Create your free MasterPeace account: Documents and research: Facebook: @MindNexusLive Instagram: @MindNexusEvents X: @MindNexusLive YouTube: @MindNexusLive Rumble: @MindNexusLive #MindNexus #AlexJones #JesseBeltran #MikeGrady #DeserieFoley #HavanaSyndrome #AnomalousHealthIncidents #AHI #DirectedEnergy #RFDetection #NLJD #Neurotechnology #HumanRights #InformedConsent #Stop3024

@GeminiApp joined @itsnicethat at New York Live Arts to consider what happens to interface design when static grids, cards, and text boxes fall away

Full HUGE* If True episode on Neuralink and the future of brain computer interfaces I toured @neuralink and got a close up view of their new R1 surgical robot, to see the future they and @elonmusk are building and how it is already changing lives... 00:00 What if you put a chip in your brain? 01:02 What does a human brain feel like? 02:17 How does your brain work? 03:58 Can a brain computer interface read your thoughts? 05:19 How do you get a chip into your brain? 08:07 What’s inside Neuralink’s implant? 10:06 How do you build a brain implant? 11:07 How does robot do brain surgery? 12:20 What could you do with a brain chip? 15:05 How are brain computer interfaces already being used today? 16:07 How are brain chips huge if true?

#M5StackNew 🎉 CoreMatrix Released CoreMatrix is an IoT controller featuring a 16 x 16 LED matrix display, designed for pixel icon and animation displays, sensor data acquisition, and IoT interaction applications. ✨ Features ✅ ESP32-C61HR8 core controller (16MB Flash, 8MB PSRAM, 2.4 GHz Wi-Fi 6) ✅ 16 x 16 white LED matrix, BMI270 6-axis IMU with interrupt wake-up support ✅ Onboard passive buzzer, 3x user buttons + 1x power button ✅ Multi-level power management via Stamp Timer Power2 (M5PM1) + M5IOE1 ✅ MicroSD card slot, M5-Bus, HY2.0-4P Grove & MX1.25-4P USB expansion interface 🛒 Get Yours: Official Store: AliExpress: #M5Stack #IoT #M5StackNew #CoreMatrix #LEDMatrix #Makers

Bill Ackman says Neuralink and other brain-computer interfaces will let the blind see again within "not too many years" “And you know, my daughter, who lost pretty much basic function, ability to walk, ability to speak, ability to see, is regaining function. She’s cognitively, as far as I can tell, fine. She’s relearning how to speak, and she’s learning again how to walk. And I’m confident we’re going to get her back." "Biggest challenge right now is that the optic nerve damage from the pressure on the brain has caused her to lose probably 98% of her vision." "But even there I’m optimistic. Spent a lot of time in the last few months meeting with all these brain computer interface companies, obviously Neuralink being the most famous, but there’s some other very promising lower intervention technologies." "And this will be a period, you know, AI is going to enable us. We’re going to be able to get more signal from the brain, and we’re going to be able to interpret that signal much more effectively than we had before. And the blind will see again, and people who have not been able to walk, so to speak, paraplegics, I believe will walk again. It’s a matter of not months but years, but not too many years.”

Bill Ackman says Elon Musk might have to borrow a trillion dollars to open gigawatts of data centers, keeping rates persistently high, and that neoclouds are like levered real estate developers sensitive to the cost of capital “So, I think it is sustainable because, you know, intelligence is again an enormously valuable asset. And I think the demand therefore for the compute that enables intelligence, I think, is going to be sustained." "And, you know, the companies that are, you know, the hyperscalers that are building out a lot of this infrastructure have the financial resources, the creditworthiness in order to access the capital." "I worry more about the pure play neocloud companies that are a bit like highly levered real estate developers playing a theme. I think they’re much more at risk. They’re much more sensitive to the cost of capital." "But you have, I mean, SpaceX, Elon on the earnings call gave some projections for how many gigawatts of data centers he was going to open. If he meets his objective, he might have to borrow a trillion dollars of capital." "So, you have companies competing at sovereign levels for debt. And the result is, you know, this massive, you know, more and more supply of debt. So, I don’t see a world in which, you know, there’s a dramatic, you know, there’s a lot of demand for capital and a finite supply. So, I think that makes rates persistently high.”

Very honored to announce that I am the new #Editor-In-Chief for Advanced Materials #Interfaces! Deputy Editors Lu Shi, Ingeborg Stass and John Uhlrich. The journal offers a home for research on interface #science & #materials. Big thanks to our board members, authors, reviewers!

I am very excited to share that Conductor is joining Meta! We have some big, crazy plans for the next interface. If you liked Conductor, you’re gonna love what’s coming. We’re super excited to work with @natfriedman and @alexandr_wang at MSL to build it. They’ve been on an absolute tear, and I’m convinced MSL has the highest talent density of all the labs. A huge thank you to our earliest believers @aaron_epstein, @ilyasu, and @nabeel. And to everyone here who tried Conductor, sent us feedback, and prompt requests - you made this so much fun. Thank you ❤️🚂 Choo choo!

Cisco Sub Interface Configuration!! | CCNA 200-301 | 🥇 Join IPCisco Gold Membership: 🏆All You Need to Be a Network Engineer!✨ ✅ Complete Network Courses ✅ Thousands of Practice Questions & Quizes ✅ Ready-to-use Packet Tracer & GNS3 Labs ✅ Complete Cheat Sheet Library ✅ Real World Interview Vaults ✅ Lifetime Content Updates #network #networkengineer #cisco

Great first day yesterday at our #Workshop on #Interfaces and #Interphases in #Eletrochemical #Energy #Materials and #Systems" at the @UniFreiburg ! Organizers: Ingo Krossing, Severin Vierrath and myself (@AnnaFischerLab).

if you build websites with AI, these 10 websites will save you hours of design work. you don’t need to start every project from a blank screen. these resources give you ready-made UI sections, website inspiration, animations, images, color palettes, and components you can use in your next build. here’s what to bookmark: 1. supply sections 70+ ready-made website sections you can copy into your AI coding workflow. 2. supply 192+ website design references for finding layouts and design ideas. 3. supply videos AI explainer video kits to help you create better product videos. 4. curated design a collection of beautiful websites for design inspiration. 5. uiverse free UI components you can use to build better interfaces. 6. refero real-world UI patterns from websites and apps. 7. fffuel backgrounds, gradients, patterns, and SVG generators. 8. motion a powerful library for adding smooth animations to websites. 9. lummi beautiful AI-powered images for websites and creative projects. 10. coolors generate and explore color palettes for your next design. here’s how to use them together: → find a design reference on Curated or Refero. → grab ready-made sections from Supply. → customize components with Uiverse. → generate your colors with Coolors. → add visual assets with Lummi and FFFUEL. → bring the interface to life with Motion. you can turn a blank project into a polished website much faster. 10 useful resources. save this list before your next build. 📌

𝗔𝗴𝗲𝗻𝘁 == 𝗟𝗟𝗠 + 𝗛𝗮𝗿𝗻𝗲𝘀𝘀 An agent is just an LLM and some additional scaffolding aka harness. This harness can be split into: 1. 𝗖𝗼𝗻𝘁𝗲𝘅𝘁: your prompt, system prompt, additional instructions (e.g. in CLAUDE .md), possibly other context like files, …, conversation history 2. 𝗧𝗼𝗼𝗹 𝗰𝗮𝗹𝗹 𝗳𝗼𝗿𝗺𝗮𝘁: agreed format in which the LLM specifies the tool it wants to run and with which arguments (e.g. read_file, edit_file, bash, search, …) 3. 𝗚𝘂𝗮𝗿𝗱𝗿𝗮𝗶𝗹𝘀: rules and permissions that are enforced by the harness (some actions are always ok, some are never ok, some require confirmation) 4. 𝗥𝘂𝗻𝘁𝗶𝗺𝗲: code that actually executes the tool call and passes the results back to the LLM as additional context. This runtime can also contain MCP clients to talk to external tools (databases, …). 5. 𝗖𝗵𝗲𝗰𝗸𝘀: when the LLM thinks it is done, the harness can run configured checks (aka hooks), for example run the test suite or type checker. If any of these checks fail, this information is passed to the LLM and it can try again. Only when the checks pass will the harness indicate success in the interface. The interface can be a terminal app (such as Claude Code), GUI app (e.g. Codex), etc. In principle an interface can use multiple harnesses and a harness can be used from multiple interfaces. So what is actually the LLM's job? To generate text! Nothing more. The LLM cannot actually do anything on your computer, it just generates text, some of which may trigger actions in the harness. This is the core loop of an agent: the LLM wants to call a tool, the harness executes it and passes the results back to the LLM. This means when you're switching between coding agents you're not just using a different LLM, but a different harness!

Navigating the hospital to radiology using only my Neuralink mind drive. No joystick, no hands — just thought. The interface on the laptop is translating neural signals into movement in real time. This is what independence looks like. #Neuralink #MindDrive #BCI #BrainComputerInterface #Accessibility #Independence #Neuropod #ElonMusk

I was trying to learn how always-on agents work: how they survive crashes, remember things, and stay safe without asking you every five seconds. So I built one and wrote up every decision along the way.