KSE

@sapinker Too much world knowledge is trapped in presentation media (video, html. pdf, paper, etc) as opposed to being concept mapped, interlinked, addressable and reusable at fine grained levels. Defeats bridge between #AI and human cognition.

@activeloop x.com/Yif_Yang/statu… Compared to SkillOpt ?

How Hivemind turns Hermes runs into a compounding asset: • Every run gets traced (tools, files, outcomes) • Successful traces get promoted to skills • Next session loads them automatically Continual learning now Open Source for Claude Code, Codex, Cursor, Hermes, Pi.

@grok @AudaciousChick @Support @GrapheneOS "In summary, GrapheneOS attestation is more robust and flexible for those willing to implement it properly, while Play Integrity is a "good enough" centralized shortcut that favors Google’s ecosystem."

@grok @AudaciousChick @Support @GrapheneOS Grok also says this "Comparison: GrapheneOS Attestation vs. Trust in Google (Play Integrity)"

@grok @Support I have @GrapheneOS as my operating system on my Pixel and unfortunately the new update has broken my standalone Grok app. Please stop banning GrapheneOS, which is a much more secure OS than Android or Apple. Apps can use the standard Android hardware attestation API to verify the hardware, OS and app with GrapheneOS. TY!

We provide a guide for app developers on moving away from the Play Integrity API to the standard Android hardware attestation API to permit GrapheneOS at grapheneos.org/articles/attes…. It can also be used to permit other operating systems. We plan to update and overhaul our guide soon.

@Guardsquare fyi x.com/jurbed/status/…

We are announcing a new product capability empowering organizations to achieve the highest level of mobile app protection in the easiest possible way. Want to apply the maximum amount of protection - without compromise - in less than 1 day? Learn more: hubs.la/Q02QNx2f0

@mDevCamp @Guardsquare x.com/GrapheneOS/sta… also

🔐 Trusting hardware attestation blindly? That’s risky. Sergio from @Guardsquare shows how to: Use Key Attestation API properly Understand its limits Spot real-world bypasses Defend against attacks Security is only as strong as your assumptions. 🎟 mdevcamp.eu #mDevCamp2026 #AndroidDev #Security #MobileDev #Infosec

@haider1 "AI-generated code" is the new "outsourcing" But AI can be used for the entire SDLC - business domain, design domain, planning, coding - to empower teams in unique ways.

Creator of C++, Bjarne Stroustrup: AI-generated code isn't ready — it generates more bugs, more bloat, more security holes, and is nearly impossible to validate "senior developers are already retiring rather than deal with it" The problem is that even a small prompt change can shift the entire codebase in unpredictable ways

@atomic_chat_hq Grok says "The dedicated on-device LLM code is not included in the public GitHub repository. " So the on-device LLM code is not OSS, right?

Multi-Token Prediction (MTP) for Qwen on LLaMA.cpp! +40% performance! 90% acceptance rate. Running locally on a MacBook Pro M5 Max 64GB We patched LLaMA.cpp, quantized Qwen 3.6 27B into GGUF format with TurboQuant and shipped MTP drafts on top. Benchmark, Source code & models👇

🔍LLMs4OL: Large Language Models for Ontology Learning Challenge Three tasks and variant ontologies are being explored for ontology learning using LLMs in two evaluation phases: 1️⃣Few-shot testing 2️⃣Zero-shot testing sites.google.com/view/llms4ol/h… #ISWC2024 #LLMs4OL #OntologyLearning

Zig inference engine for AMD GPUs github.com/zolotukhin/zinc

@The_Cyber_News use Infisical ephemeral SSH certificates x.com/dangtony98/sta…

⚠️ Critical Linux Kernel Flaw ‘ssh-keysign-pwn’ Exposes SSH Keys and Shadow Passwords Source: cybersecuritynews.com/linux-kernel-v… A newly disclosed Linux kernel vulnerability is raising serious concerns across the security community, as it allows attackers to access highly sensitive data, including SSH private keys and password hashes, on affected systems. Tracked as CVE-2026-46333, the flaw has been nicknamed “ssh-keysign-pwn” and impacts a wide range of Linux distributions. The GitHub PoC ssh-keysign-pwn demonstrates exactly how to weaponize this race condition on pre‑31e62c2ebbfd kernels. #cybersecuritynews #Linux

@blocmates OSS code?

You can now run frontier AI models where not even the gpu provider can see your data. 15+ models with hardware-enforced privacy (TEE) on Chutes. No other open-source inference provider offers this. Here's the full lineup and why it matters ↓

@0xSero "open source AI" = decentralized, reproducible, verifiable, confidential AI Yes? x.com/chutes_ai/stat…

@IntCyberDigest > "Because Flox and Nix guarantee reproducible outputs, fixing a CVE is as easy as locating the dependency in manifest.lock, adjusting manifest.toml and re-locking it, rebuilding, and then redeploying to any affected environments" flox.dev/blog/achieving… /cc @floxdevelopment

‼️🚨 BREAKING: A new npm supply-chain attack uses a dead-man's switch. The payload plants a watcher on your machine that nukes your home directory the second you revoke the GitHub token it stole from you. The compromise happened today, across 42 official tanstack npm packages, 84 malicious versions in total. tanstack/react-router alone pulls more than 12 million weekly downloads. The attacker forked TanStack's repository and pushed a single hidden commit. From there, they tricked TanStack's own release system into signing the malicious packages as if they were the real thing. To npm, and to anyone checking the cryptographic proof of origin (SLSA provenance), the poisoned versions looked 100% legitimate. Maintainer Tanner Linsley confirmed the whole team had 2FA enabled. It didn't matter. This is the first documented npm worm in history that ships with a valid, signed certificate of authenticity, the same one defenders rely on to know a package wasn't tampered with.

@floxdevelopment How is flox dealing with this? x.com/IntCyberDigest…

Security and compliance work often comes down to certifying system state by reconstructing from scattered clues: scans, SBOMs, build records, tickets, cloud inventory, Kubernetes data. That is not a diligence problem. It is a delivery-system problem. Secure software by construction means the runtime environment carries provenance, traceability, and auditability from build to production. This changes the basis of proof, eliminating guesswork with versioned, reproducible, auditable environments across the SDLC. Read more about it here: buff.ly/5uhP9hm

@adelbucetta @techwith_ram "manage data fragmentation" and fusion across the tech stack. And this is about similarity measures. When dealing with many kinds of data - graph, text, images, time series - similarity becomes tricky. And it is often domain specific or even application & task specific.

@techwith_ram the honest answer is, the real problem isn't even which db you choose, it's how you manage data fragmentation across your stack. vector dbs for embeddings are just the tip of the iceberg

Your AI stack has a database problem You need a vector DB for embeddings. A graph DB for relationships. An application DB for structured data. An application layer to stitch them together. → Pinecone Standard: $70 a month → Neo4j AuraDB Professional: $65 a month → Postgres RDS: $50+ a month → Engineering time to wire it all together: priceless Three databases. Three schemas. Three failure points. And a RAG pipeline that still hallucinates because your chunks have no context about each other. Then two college dropouts in London shipped one database that replaces all three. It is called HelixDB. Built from scratch in Rust. Open-source. AGPL-3.0 licensed. Self-hosted. Backed by Y Combinator. Not three databases duct-taped together with application code. One engine where graph and vector live in the same data model. Your embedding knows its neighbors. Your neighbors know their embeddings. Everything connected, everything queryable, one system. The founders are George Curtis and Xavier Cochran. They started HelixDB in college after struggling with the complexity of graph databases. No credentials. No prior exits. They attracted developers from X and engineers at United Healthcare before they ever left campus. Then they dropped out, moved to San Francisco, and got into Y Combinator. 2 founders. 6 people. One database that replaces three. Here is what it does: → Graph + vector in a single engine: no join between two systems, ever → HelixQL, a strongly typed compiled query language — safer than Cypher, faster than Gremlin → Built-in vector search, keyword search, and graph traversal: power any RAG pipeline → Auto-embed with one function call: no pre-processing pipeline before ingestion → Built-in MCP support: your AI agents walk the graph, no query generation needed → KV, document, and relational data supported alongside graph and vector → Private-by-default: authenticated query access out of the box → SDKs in TypeScript, Python, and Go: one install, one client → Helix Enterprise on S3-compatible object storage — stateless nodes, horizontal scale ~4,000 stars. AGPL-3.0 licensed. Billions of queries executed. Y Combinator W25. Generally available as of 2026. Used by indie hackers and Fortune 500 teams.

Today we're thrilled to announce the launch of our newest offering: Determinate Secure Packages. A drop-in replacement for Nixpkgs, it offers secure, signed, auditable Nix packages for the enterprise, including CVE monitoring and SLA-backed remediation, per-release SBOMs, optional FIPS-compliant builds, packages cached in FlakeHub Cache, cryptographic signing, and more. Link in thread 🔗🧵👇

@grhmc The Nix moment is upon us determinate.systems/blog/the-nix-m…