Operational databases have long relied on tightly coupled compute... and storage. This architecture creates resource contention and pushes teams to manage infrastructure rather than build. As applications become more real time and automated, the transactional layer needs to adapt. Databricks Lakebase is built for that evolution: • Familiar Postgres semantics for app developers • Compute separated from durable state • Operational data running directly on the lakehouse • Serverless autoscaling (including scale to zero), branching, and recovery to match agent-driven workload Now generally available:show more

What Postgres can’t do alone, it can tackle with... DuckDB. The pg_duckdb extension shows how two databases can work hand in hand by querying analytical data with DuckDB and joining it with transactional data in Postgres. The extension embeds DuckDB directly into Postgres. DuckDB literally runs inside a Postgres process. How it works (see diagram): 1. An app sends a query to Postgres joining portfolio data (Postgres) with historical market data (lakehouse) 2. pg_duckdb detects lakehouse access and delegates execution to DuckDB 3. DuckDB pulls required market data from the lakehouse, queries Postgres tables, and performs the join 4. Postgres returns the final result to the appshow more

Secure, programmable key management infrastructure for developers building on... Arc! Own the wallet experience, control what gets signed, and handle operational flows without the complexity. The infra to build at scale, with the security to match.show more

Partnership Announcement: Exabits x Mizzle, Powering AI Compute with... Web3 Innovation Exabits and Mizzle are teaming up to redefine AI infrastructure and unlock new opportunities in AI compute. Exabits – The leading AI compute network, tokenizing GPUs to unlock real yield and make high-performance infrastructure accessible and monetizable. – The Web3-native marketplace streamlining tokenized AI services for seamless adoption and scalability. Shared Mission: Breaking barriers in AI compute, making high-performance, advanced infrastructure accessible, monetizable, and profitable for everyone. Why this partnership matters: - Seamless GPU Liquidity – Exabits tokenizes AI compute, Mizzle creates frictionless access. - On-Demand AI Power – Institutions & developers instantly access scalable compute resources. - Transforming AI Infrastructure – Making high-performance compute an asset class, fueling next-gen applications. AI compute is evolving. Real compute. Real adoption. Real revenue. This is how Web3 wins.show more

🔐 𝗣𝗼𝘄𝗲𝗿𝗶𝗻𝗴 𝘁𝗵𝗲 𝗡𝗲𝘅𝘁 𝗪𝗮𝘃𝗲 𝗼𝗳 𝗔𝗜 We're proud... to partner with Stempoint to secure the future of decentralized AI compute. StemPoint is building one of the most advanced distributed AI infrastructures on the planet, sharded GPU compute, open model libraries, and secure data flows across a global mesh. Now with Naoris Protocol’s Sub-Zero Layer and dPoSec consensus, every node, model, and data packet gains: • Post-quantum cryptographic protection • Real-time validation of compute and algorithms • A decentralized trust layer for sovereign AI 🚫 No more trusting black boxes. ✅ Verifiable AI infrastructure at scale.show more

Behind every OptimAI Node is a mission far greater... than rewards: 🔸To build a decentralized Reinforcement Data Network 🔸To unlock Agentic AI for everyone—not just a privileged few 👉OptimAI Edge Node: Our architecture is now in motion: 🔸OptimAI DePIN to power decentralized infrastructure 🔸OptimAI DeHIN to amplify collective intelligence 🔸Reinforcement Data Layer to train agents smarter, fairer 🔸Compute Layer to enable real-time AI at the edge 🔸OptimAI Chain to govern everything, transparently and efficiently OptimAI Agent Studio, Agent OS, Data Engine, Compute Engine: all part of what we’re building next. This is a network built by people, for people. And you’re not just early, you’re essential. Let’s keep mining, contributing, earning rewards in return, and rewriting the future of Decentralized AI. The future is Agentic. The fuel is your data. The power is decentralized. One Node. One Data. One Agent at a Time. #BUIDL with us!show more

Data Streaming 🤝 Data Intelligence Confluent and Databricks are... teaming up to empower organizations with AI-ready, real-time data. Our seamless integrations between Tableflow, Delta Tables, and Unity Catalog will connect operational and analytical systems, to power AI-driven applications. Learn how we’re bridging the data divide in Jay Kreps and Ali Ghodsi’s blog ➡️show more

Introducing Databricks LakeFlow, a new solution that simplifies all... aspects of data engineering through a simple, unified & intelligent experience. Easily build production-grade data pipelines to scale to the ever-growing demand for reliable #data & #AI.show more

Replit, Vercel, and OpenAI have built very cool agent-native... applications, but nobody else has passed the demo stage. Building agents that work is complex. Teams aren't shipping agents because we don't have good tooling yet (and most of us don't know how to do this well.) A couple of days ago, the CopilotKit🪁 team announced a collaboration with . You can now use LangGraph with CoAgents to build agent-native applications, and here is everything you need to know about that: CoAgents is fully open-source, and you can use it to do the following: • Human-in-the-loop to steer and correct the agent • Stream intermediate agent state • Real-time state sharing between the agent and the application • Agentic generative UI to build trust that the agent is on the right path Start this GitHub Repository: Thanks to the team for giving me early access and collaborating with me on this post.show more

NVIDIA might have just declared war on the cloud... GPU business For years, AI builders had one option Rent compute Pay every month Watch the bill grow every time usage increased Now NVIDIA is putting serious AI hardware directly on people's desks Small enough to fit next to a monitor Powerful enough to run workloads that used to require expensive cloud infrastructure That's why this launch is getting so much attention The real story isn't the hardware specs It's the business model shift Every month, developers send money to cloud providers for inference, testing, fine-tuning and AI applications The question nobody can answer yet is what happens if enough developers decide they'd rather buy infrastructure once than rent it forever Because if local AI hardware keeps getting more powerful, the economics start changing very quickly Cloud providers built empires on renting access to compute NVIDIA is betting more people will eventually want to own it And that's a much bigger story than a new piece of hardware sitting on a deskshow more

Aethir welcomes Chainlink to the AI Unbundled alliance ⛓️‍🌐... With Chainlink joining the alliance, devs will now have access to both premium GPU compute from Aethir and the trusted data layer (Chainlink) to build scalable, decentralized AI apps. Read more 🧵👇show more

⚙️ We’re nearing the completion of the Edge Unified... API, a single decentralized interface to access the full power of Edge. | GPUs, Compute, Storage, AI, and more. Built for the future, where robots, agents, and applications tap directly into the Supercloud. Stay tuned! 🔜🦿show more

March 18, 2025 marked the public launch of OptimAI.... In one year, it has evolved from a lightweight node layer into a decentralized intelligence infrastructure powering real-time data, compute, and reinforcement for agentic systems. Not just nodes. Not just data. A continuously learning, network-driven intelligence layer. This is infrastructure for a new class of software: autonomous agents that persist, adapt, and operate across environments. Year one established the network. Year two is where it compounds into coordination and value flow. Personal agents. Reinforcement at network scale. Emerging primitives for AgentFi. New layers coming online. 2026 won’t just be about scale, it’s where the network starts to operate. Keep building!show more

The future runs on a unified compute foundation. Adapt... faster across generations and locations with a compute platform built for whatever comes next.show more

Weekend read: 𝗔𝗴𝗲𝗻𝘁𝘀 𝘀𝗵𝗼𝘂𝗹𝗱 𝗯𝗲 𝗦𝗲𝗿𝘃𝗲𝗿𝗹𝗲𝘀𝘀 (𝗮𝗻𝗱 𝗗𝘂𝗿𝗮𝗯𝗹𝗲) Serverless... platforms are amazing for many AI agents, especially if they do a lot of bursty work (chat sessions with long inactivity) or react to events (approvals). Two things are needed to make this work well: (1) Durable execution / orchestration, to let agents recover and suspend (scale to zero) when they wait for input. (2) A durable execution orchestrator that has a push model that works with serverless platforms. Super fun to see agents with restate go brrrrrrrrrr: starting as durable functions on Modal's compute platform, scaling up, running, scaling down. Just out of the box, no extra code, no workers deployed. 👇show more

One of the largest anti-encroachment drives on illegally occupied... Railway land near Bandra station is now underway. This long-standing issue had created serious safety and operational challenges repeated trespassing, dangerous track crossings, risk to lives, disruption to train operations, and roadblocks to critical infrastructure upgrades. This railway area stretch is operationally critical for augmentation of rail capacity in Mumbai. Removal of encroachments is essential for railway safety, operational flexibility and future infrastructure expansion, including additional train services. #WesternRailway #BandraStationshow more

Other than Arweave, I’m struggling to think of an... endpoint an agent can hit to store data, for free, in one http call -- let alone store it permanently. For compute, most solutions are rented, require human intervention to set up, or do not stack up well against what’s available from web2. Agents cannot effectively acquire their own infrastructure for memory and compute - and what they end up acquiring will not be magically everlasting. The way we go from localhost to production-scale deployments still needs humans. It needs to be easier than asking their human to sign up for AWS and add a credit card. For every new VM, it’s configuring packages, ensuring the services spawn and stay alive, remembering how to access it, maintaining a map of the infra... It’s clear that crypto wallets are how we make it frictionless for agents to provision resources autonomously, but for permanent compute, unkillable processes, high availability and no API keys, we’re largely stuck in the sandboxed realm of EVM/SVM. And even then, many of the things that make these onchain compute layers usable in production depend on paid services like indexers and RPCs. This is why I see the permaweb as the ideal compute substrate for agents. ao processes are not limited to the EVM sandbox. They can effectively do anything a web2 serverless functions platform can do. And all the agent needs to do is generate a wallet with arweave.js and push the process. No human intervention needed. Of course, there’s a knowledge gap -- models know about Arweave and will choose it more often than not when asked to store data permanently, but they often fall back to legacy patterns. PierreSteinClaysky 🐘 states this here along with some deeper probes into how proficient Codex is at modern permaweb development: Until the models catch up -- and the first doc they find is the right doc -- you can put them on rails. A RAG/text database for the new best practices of the permaweb is available here: Agents are building a new world while we sleep. We’d better make sure it’s not on vaporware foundations.show more

Vibe coding apps that require you to store and... manage large files just got easier, with Replit ⠕ App Storage. ✨ My favorite example in this category are any PDF analysis apps: create a fully customized AI synthesizer in just a few prompts. App Storage is Replit's Object Storage solution (like Amazon S3 or Google CS). But don't worry if those words sound foreign to you, Agent will set this up for you automatically if your app needs it. On that note, this Agent capability is just the beginning of our efforts to make building apps with databases and storage easier: 👇show more

Partnership Announcement Exabits x {QuillAI Network} : AI Compute... Meets Autonomous Security The future of Web3 doesn’t just run on AI—it protects itself with it. Exabits and QuillAI Network are teaming up to push high-performance AI compute + adversarial AI security into the onchain economy. Who’s in the Room? Exabits → The Mothership of AI Compute: The Base Layer Everyone Builds On, scaling autonomous intelligence at the edge. QuillAI Network → AGI-grade security, where adversarial AI agents run trustless audits on smart contracts. What’s Happening? We’re cutting the friction between AI infrastructure and Web3 security. AI models need compute; smart contracts need protection. Now, they get both. Secure AI Compute: Exabits fuels QuillAI’s adversarial AI, giving security agents the firepower to scale autonomously. AI-Powered Defense: QuillAI safeguards Exabits’ compute layer, ensuring models run without vulnerabilities. Smarter Web3 Infrastructure: The merge of AI compute + security agents unlocks trustless AI governance, fortified DeFi stacks, and scalable digital ownership models. What’s Next? Web3 isn’t waiting. AI autonomy needs distributed compute + bulletproof security to thrive—and we’re delivering it. This partnership sets the foundation for AI systems that build, protect, and scale themselves. The new AI economy isn’t coming—it’s already here. Let’s move.show more

We're excited to announce two new products, the Grayscale... DeepBook and Walrus Trusts $DEEP $WAL. Built on Sui, $WAL helps enable decentralized data management at scale, while $DEEP serves as the foundational liquidity layer for DeFi applications. 👇show more

Introducing the Grayscale Walrus Trust $WAL 🧊 Walrus 🦭/acc... is redefining decentralized data management, a critical pillar for scaling real-world Web3 applications. 🧠 Programmable Data at Scale: Walrus enables on-chain storage and real-time data management, powering dApps like social networks, games, and productivity tools — all while maintaining privacy and ownership. 🔗 Chain-Agnostic by Design: Walrus can store, share, and manage data across any blockchain (not just Sui), giving developers flexibility and speed. 🏗️ Core Infrastructure: From encrypted file sharing to active deletion and lifecycle controls, Walrus is a foundational piece of Mysten Labs’ decentralized stack. See important disclosures and learn more about the Grayscale Walrus Trust:show more