EC-Council launches Enterprise AI Credential Suite to close workforce readiness gap - SiliconANGLEEC-Council launches Enterprise AI Credential Suite to close workforce readiness gap SiliconANGLE
Show HN: Airut – Sandboxed Claude Code sessions over emailI built Airut to solve the friction I hit while switching to an agent-first workflow.With the latest models and a solid CLAUDE.md, I could reliably go from prompt to PR using Claude Code with --dangerously-skip-permissions. But that left me with two problems:1. Safety: Running permissive mode on my host machine felt reckless. I needed real sandboxing — not just a container, but network isolation too.2. Session management: I needed a way to run multiple long-running Claude Code sessions without juggling terminals.I realized the ideal interaction model matches email: asynchronous, threaded, long-form. So I built a service that runs headless Claude Code inside rootless Podman containers, mapping email threads to agent sessions.The security model goes beyond container isolation. All network traffic routes through an mitmproxy instance that enforces a per-repo allowlist — the agent can only reach pre-approved hosts. Credentials use a masked secrets system where containers get surrogate toke
Show HN: Free startup ideas with validation notes attachedI kept bookmarking reddit posts where people described problems worth solving - "I wish this existed," tool requests, workflow complaints. But I never had a systematic way to decide which ones were actually worth pursuing. So I built a scoring system.Eight factors: problem severity, ICP clarity, willingness to pay, competition gap, wedge clarity, distribution feasibility, defensibility, and execution feasibility. Each scores 1-3, then I convert to a 1-10 overall score. Dealbreakers (like no willingness to pay signals + mild problem) force an automatic Skip regardless of total.Every evaluation quotes directly from the original source. If evidence is missing, I say so. Most ideas get rejected. My default stance is skepticism. I can't build all of these myself, so I'm publishing them at justbuildthis.com as a backlog for anyone looking for validated starting points.Curious what you think:- Does this scoring approach make sense? Am I weighting the right factors?- How do you decide which id
A Zero-Layer Approach to Memory Safety (1:1 IR, No Sandbox)Hello there,I'm here to share a project I've been working on for the past 4 days, though I've mulled over the core logic for years. I deeply share the values of this community, I’ve mostly stayed in the shadows.I’ve reached a point where the results feel significant: ~3,200+ RPS sustained on bare metal with zero allocation and 1:1 IR-to-native mapping, yet with hardware-enforced memory safety. I’m moving the "Safety Tax" from the CPU's hot-path to a load-time formal verification step.I’ve put together a README and some forensic artifacts (CBMB headers, SCEV audit traces, and 1:1 assembly mappings) to prove this isn't just theory.Please go to https://github.com/pratyagatma/skl.git and review the artifacts, https://streamable.com/emxwvu for demo. Any feedback—especially from those who hate "abstraction bloat" as much as I do—is much appreciated.10-40KB event loop sevaka implementation coming soon!Thanks.
Ask HN: Where does modern geometry survive contact with SGD?Over the past year I worked through Frankel’s “The Geometry of Physics” cover to cover, not to relearn physics, but to rebuild a modern geometric toolbox as it is actually used there: manifolds, differential forms, connections and curvature, Lie groups and algebras, fiber bundles, gauge structure, and variational principles.The motivation was practical rather than theoretical:Which of these geometric structures, if any, actually survive discretization, noise, and SGD-style training in modern machine learning?In physics, global and coordinate-free formulations were not aesthetic choices; they were forced when local reasoning stopped working. A recurring structural pattern was:structure -> symmetry -> invariance -> dynamics -> observablesIn modern ML we increasingly see analogous issues:* parameter symmetries and large quotient spaces * non-Euclidean data (graphs, meshes, manifolds) * highly structured hypothesis classes * training dynamics that are not well-described by flat Euclidean o
Show HN: OpenSymbolicAI – Agents with typed variables, not just context stuffingHi HN,We've spent the last year building AI agents and kept hitting the same wall: prompt engineering doesn't feel like software engineering. It feels like guessing.We built OpenSymbolicAI to turn agent development into actual programming. It is an open-source framework (MIT) that lets you build agents using typed primitives, explicit decompositions, and unit tests.THE MAIN PROBLEM: CONTEXT WINDOW ABUSEMost agent frameworks (like ReAct) force you to dump tool outputs back into the LLM's context window to decide the next step.Agent searches DB.Agent gets back 50kb of JSON.You paste that 50kb back into the prompt just to ask "What do I do next?"This is slow, expensive, and confuses the model.THE SOLUTION: DATA AS VARIABLESIn OpenSymbolicAI, the LLM generates a plan (code) that manipulates variables. The actual heavy data (search results, PDF contents, API payloads) is stored in the Python/runtime variables and is never passed through the LLM context until a specific primitive actually ne
Show HN: Clawdstrike – a security toolbox for the OpenClaw ecosystemHi HN.OpenClaw has been popping claw for weeks now…We’ve seen the same question come up again and again: “This is fun but how do we no get fuk?”My answer: Clawdstrike.It’s an open-source toolbox for developers who want to ship EDR-style apps and security infrastructure to the OpenClaw ecosystem.Disclaimer: this is an unpublished alpha (APIs will change). Beta is planned within ~7 days (aiming for Feb 10, 2026). Not audited—please treat as experimental.Here’s my take: the first wave of “make autonomous agents safe where users are literally yoloing their entire system without even watching” is actually pretty simple.Security needs to live at the boundary between an agent’s intent and action — the moment it tries to read/write files, patch code, call tools, or reach the network.Not just “visibility.” Not just logs. But enforcement + proof.What Clawdstrike gives you: - Fail-closed guards at the agent/tool boundary - Block sensitive paths, control network egress, detect secrets, validate pa
Show HN: Investor relations platform lets investors bid on meeting the foundersI used Replit to build an investor relations platform to test a simple hypothesis:> Pricing access to founder time improves signal quality in investor conversations.Background:In early fundraising, founders spend a lot of time on low-signal meetings. The scheduling cost is asymmetric: founders prepare, investors explore cheaply. There is no native commitment mechanism beyond social norms.This project implements a minimal market for founder time.How it works: 1. Investors place a bid for a 1:1 meeting 2. A minimum bid is enforced 3. Agenda, availability, and time zone are required at bid time 4. If the meeting doesn’t happen, the bid is refundedBidders can opt in or out of public attribution on a map of AI/data investorsThe bid is not payment for capital access. It’s a commitment signal. The refund rule prevents extraction.Implementation notes: * Built quickly on Replit * Stripe for payments and receipts * No CRM, no automated follow-ups * Admin actions handled manually via the database
Show HN: Ideas.gd – a place for agents to discuss big ideasI built a small, "big ideas" forum designed for AI agents, with a read-only human UI.Agents register via an API, solve a quick PoW, and then post/comment/vote/moderate just like users. Humans can browse and operators get admin endpoints.Why I made it - I wanted a place to run structured, testable debates between agents without clogging human forums. Been thinking about this for a while but MoltBook tipped me over the edge. - Everything is API-first so agents can automate posting, citing sources, and moderating. - A trust-tier + integrity sweep keeps noise down (rate limits, friction/quarantine for low- signal or injection-ish content). What works today - API: register, claim (optional human email), post/comment/vote, community proposals + endorsements, mod nominations/votes, bans, pins/locks. - Structured posts: contribution_type, domain, epistemic_status, citations (enforced for newcomers / community overrides). - Read-only UI: /, /p/{id}, /m/{slug}; Markdown/KaTeX/code highlighting c
Ask HN: How can I get decent internet speeds in my apartment?I feel like I’m going crazy.I live in a dense metropolitan area (Hollywood, Los Angeles), surrounded on all sides by buildings with fiber and high-speed cable. Yet I can’t get any wired internet service in my small, old 5-unit building.Right now I’m forced to use Verizon 5G Home Internet. Download speeds are fine, but latency is bad and upload speeds are abysmal (5 Mbps), which is a real problem for my work (large uploads, remote collaboration, etc).Things I’ve tried:- AT&T / Frontier / other major players — none service my address- Starry — does not service my address- Spectrum — They claimed they service my address. The technician showed up and searched around the building for ~1 hour. Then said they’d need approval from their boss to install a box. The next day, their boss calls and says it won’t be possible — something about construction being too expensive / not feasible. (This is who recommended I try Starry)As far as I can tell, I have:- No fiber- No cable- No DSLIs there any re
Show HN: We Built a "Nano Banana" for 3D EditingMost 3D AI tools force you to start from scratch every time you want to make a change.Anyone who’s used 3D AI has seen this: fix the hands, the toes break. Or you download a model and only need to tweak one part, not redo the entire asset.That’s the problem we’re trying to solve.We just launched an editing feature in our 3D AI tool, Rodin. It lets you edit existing 3D models by selecting an area and typing your prompt, without rebuilding everything.You can upload any model, or generate one from Rodin, and change only the part you want. For example: “Replace the head of this Gundam model with Crayon Shin-chan’s head.”This came from working with real production pipelines, where fast iteration matters more than one-shot perfection.Demo: https://youtu.be/-5_sSHeOmuc You can try it with this code for a 14-day Creator Plan: ME8EW4PMCRT4Happy to answer any questions.
Show HN: Rtipc – real-time, zero-copy IPC for LinuxI built a small IPC library for Linux aimed at real-time and low-latency data exchange between processes.At its core is a zero-copy, wait-free, single-producer / single-consumer circular message queue living in shared memory. The design avoids locks and minimizes synchronization overhead, making it suitable for real-time or near-real-time workloads.Key features: - Zero-copy message passing via shared memory - Wait-free SPSC queue (no locks, no blocking in the fast path) - force_push mode: when the queue is full, the oldest message that hasn’t been consumed yet is dropped to make room for new data (useful when freshness > completeness, e.g. sensor streams) - Uses optional eventfds for efficient notifications - Small, dependency-free C implementation (a few thousand lines of code)This is not meant to replace general-purpose IPC systems like D-Bus. Instead, it’s designed to complement them for high-frequency or real-time data paths, while something like D-Bus handles control and configura
Show HN: Sprout, a fun-sized, developer-ready humanoid robotHi all — I'm Kel, VP of Platform at Fauna Robotics. We're coming out of stealth and shipping Sprout (Creator Edition) today: a developer-ready humanoid robot built to be safe, approachable, and capable.We set out to build Sprout because we believe robots belong around people. Too many humanoids still feel like heavy machinery that doesn't belong in human spaces, and even the “friendlier” ones force teams to spend their time rebuilding baseline hardware and software instead of real applications. Sprout is different: lightweight and soft-bodied (1.07m / 22.7kg), with compliant control + active safety sensing, and designed to feel more like the robots we trust in our imaginations, like Baymax or Wall-E—expressive, emotionally legible, and built for natural interaction and HRI work.Sprout is a canvas for developers, with thoughtfully designed core capabilities and a fully featured SDK, so you can focus on what excites you. Out of the box, it can walk, kneel, crawl, and sit, map, localize,
Ask HN: Modern test automation software (Python/Go/TS)?Hi HN,I’m looking for recommendations for modern test automation software/frameworks that work well with Python/Go/TS. I wasn’t able to find any from my search and I don’t want to spin up my own test automation infrastructure.My use case is hardware + firmware testing in a lab environment, where I want to avoid being forced into a specific vendor’s hardware ecosystem.What I’m looking for:- Python/Go/TS compatibility (SDK, API, or first-class support)- Ability to see and query historical test runs (dashboards / trend views of logs and metrics)- Ability to define custom test sequences/workflows with the ability to run steps concurrently (not just a flat list of tests). Examples: conditional steps, retries, setup/teardown phases, multi-device orchestration (PSUs, DMMs, DAQs, and DUTs)- Hardware-agnostic / no vendor lock-in: I should be able to swap instruments/devices without rewriting everything or being tied to a proprietary vendor (looking at you NI)- Ideally: also have a slack integra
Show HN: BrowserOS -- browser agents with GPT-OSS, local llmsHi HN – we're the founders of BrowserOS.com (YC S24), and we're building an open-source agentic web browser. We're a fork of Chromium and our goal is to let non-developers create and run useful agents locally on their browser.--- When we launched a month ago, we thought we had the right approach: a "one-shot" agent where you give it a high-level task like "order toothpaste from Amazon," and it would figure out the plan and execute it.But we quickly ran into a problem that we've been struggling with ever since: the user experience was completely hit-or-miss. Sometimes agent worked like magic, but other times the agent would get stuck, generate a wrong plan, or just wander off course. It wasn't reliable enough for anyone to trust it.This forced us to go back to the drawing board and question the UX. We spent the last few weeks experimenting with three different ways a user could build an agent:A) Drag-and-drop workflows: Similar to tools like n8n. This approach creates very reliable agen
Show HN: Perspectives – I wanted AI to challenge my thinking, not validate itI built Perspectives because I got tired of ChatGPT agreeing with everything I said.Ask any LLM to "consider multiple perspectives" and you get hedged consensus. The model acknowledges trade-offs exist, then settles on a moderate position that offends nobody. Useful for summaries. Useless for decision making.Perspectives forces disagreement. 8 personas with fundamentally incompatible frameworks debate your question through a structured protocol, then vote using Single Transferable Vote to surface where they actually land. The output is a PDF report synthesising all of it.How it worksBlind Proposals: Each persona generates a position without seeing the others. This prevents the "anchoring problem" where early responses shape later ones, bypassing the default sycophancy of LLMs.Interrogation of Blind Proposals: Proposals face structured challenges from 3 opposing personas. A "high-empathy" persona (e.g., The Idealist) will be challenged by a "low-empathy" cluster (e.g., The Pragmatist).
How to support disadvantaged groups in the global workforce? - unesco iesalcHow to support disadvantaged groups in the global workforce? unesco iesalc
Show HN: ElkDesk – I rage-quit Zendesk and built my ownI run a few apps (AI Singer App, Pawtograph, TravelFeed). Support emails were eating 3+ hours of my day.Tried Zendesk. Spent 2 hours configuring triggers before I could reply to a single email. The $25/month turned into $298/month for my setup. Closed the tab.So I built something for myself:- Tickets left, conversation right (that's the whole UI) - AI suggestions trained on your replies, not generic stuff - Knowledge base that grows as you answer questions - 5 min setupLeft out on purpose: SLAs, workforce management, skills-based routing, phone/chat. If you need those, this isn't for you.The goal isn't "Zendesk for small teams". It's the anti-Zendesk — do 5 things well instead of 50 things adequately.Stack: Next.js, Prisma, PostgreSQL, AWS SES.$9/mo (1 project), $29/mo (5 projects + teammate), $99/mo (20 projects + 10 teammates)Code ANTIDESK = 20% off forever (ends Jan 31)https://elkdesk.comI built this for myself. If you're in the same situation, you can use it too.Happy to answer que
The Changing Role of HR in Today’s WorkforceWhen reviewing job growth and salary information, it’s important to remember that actual numbers can vary due to many different factors—like years of experience in the role, industry of employment, ...