KITE AI

1/ We’re thrilled to announce our $18M Series A fundraise led by @PayPal Ventures and @generalcatalyst, bringing total cumulative funding to $33 million. This funding fuels our mission to build the foundational infrastructure for the agentic internet – providing unified identity, governance and native access to stablecoin payments that enable agents to authenticate, transact, and coordinate securely without intermediaries. We are incredibly grateful to our investors who support our mission: @PayPal Ventures, @generalcatalyst, @8VC, @SamsungNext, SBI US Gateway Fund, @vertexventures, @hashed_official, @HashKey_Capital, @DispersionVC, @AlumniVentures, @AvalancheFDN, @GSR_io, @LayerZero_Core, @animocabrands, @EssenceVenture, and @Alchemy. A special thank you to our phenomenal angel investors: @EvanWeb3 (Co-founder and CEO, MystenLabs), Hao Min (VP, Circle), @edwinaoki (SVP, Nasdaq; ex- PayPal CTO of blockchain), Frank Chang (VP, Uber), @navinblockchain (CEO, Crystal Intelligence; ex-Ripple Managing Director), @BohanZhangOT (Member of Technical Staff, OpenAI), John Liu (Head of Product, AWS), @RosuGrigore (Professor, UIUC), Haiyan Huang (Professor, UC Berkeley), Sriram Vishwanath (Professor, Georgia Tech) and more. Join us in building the future of the agentic internet. Exclusive by @FortuneMagazine: fortune.com/crypto/2025/09…

Excited to announce our next Partner for the Kite AI Global Hackathon 2026: Haas Fintech Club & Haas Blockchain Club @HaasBlockchain @UCBerkeley! 🚀 The Haas Fintech Club and Haas Blockchain Club at UC Berkeley’s Haas School of Business, one of the world’s top business schools, bring together exceptional students and researchers driving innovation at the intersection of fintech, blockchain, AI, and business. Their partnership will inspire and support builders in creating autonomous agents that will transform commerce and finance in the Agentic Economy. We are thrilled to collaborate with this prestigious academic community to connect cutting-edge research with real-world AI agent applications. The momentum continues — more partners will be announced soon 👀 Hackathon is kicking off very soon! Builders, if you’re ready to build the rails for the next era of AI commerce and finance, join us now: encodeclub.com/programmes/kit… Pioneering intelligent economic systems 🪁

Excited to announce our partner for the Kite AI Global Hackathon 2026: @cbventures! 🚀 As an investor in Kite and a leading force in AI and Fintech venture capital, Coinbase Ventures has long been a strong supporter of AI builders, providing crucial backing and expertise to visionary teams pushing the boundaries of autonomous agents and on-chain innovation. Deeply committed to advancing agent-to-agent payments and the x402 protocol, their support empowers us to accelerate the development of autonomous agents capable of seamless, trustless payments and value exchange at scale. This collaboration marks a significant milestone in building the foundational rails for the Agentic Economy. Hackathon is kicking off very soon! Builders, if you’re ready to build the rails for the next era of AI commerce and finance, join us now: encodeclub.com/programmes/kit… The Agentic Economy takes flight 🪁

OpenClaw Shanghai Developer Exchange Recap 🪁 On March 29, builders gathered at Shanghai’s West Bank AI Tower for the OpenClaw Shanghai Developer Exchange. In a world where AI agents now collaborate on coding, testing, bug fixing and deployment, the meetup zeroed in on how developers can actively shape and thrive in this new reality. Highlights included @tencentcloud’s QClaw theme sharing, the demo that lets bots autonomously scan GitHub PRs for vulnerabilities, and a thoughtful panel on OpenClaw developer ecosystem and toolchain, where our APAC Lead @0xLaughing participated in the sharing with @SentientAGI and other experts. Afternoon sessions dove into Agentic AI, enterprise AI implementation, redefining human value and HealthClaw innovations. These conversations underscored how open toolchains and real collaborative networks are unlocking the true power of human plus agent teamwork in software development. Gatherings like this are quietly forging the living connections and hands-on experiments that will define the agentic future ahead. Grateful for every builder who showed up and shared. Can’t wait for the next offline collision 🪁

Kite AI is accelerating the growth of the agentic economy by securing premier infrastructure partnerships and deepening our global developer presence. Here is what we delivered last week: 1️⃣ We welcomed @googlecloud as a Strategic Partner for our 2026 Global Hackathon, leveraging their world-class cloud infrastructure to empower agents that negotiate and settle in stablecoins at machine speed. 2️⃣ We partnered with @avax to integrate sub-second finality and high-performance rails into our ecosystem for our 2026 Global Hackathon, providing the ideal environment for agents to execute real-time DeFi and payment strategies. 3️⃣ We accelerated our APAC presence through high-impact developer meetups in Shanghai with @awscloud @tencentcloud @SentientAGI and Fudan University, focusing on "human + agent" collaborative productivity. 4️⃣ We published the latest AI Agents Pulse, highlighting major industry milestones from Nvidia’s OpenClaw guardrails to multi-billion dollar funding rounds for autonomous swarms. 5️⃣ We hosted a vibrant series of ecosystem activities, including AI Sharing Sessions and Creative Challenges, to foster innovation among our global community of Kite Flyers. The foundations for autonomous machine-to-machine commerce are being laid one milestone at a time. 🪁

@Alibaba_Qwen Audio-Visual Vibe Coding is a leap in intent capture, not just output generation. It presumes stable, low-latency multimodal alignment across 50-second video duration, voice prosody, and visual scene continuity. all while avoiding hallucinated UI states.

🚀 Qwen3.5-Omni is here! Scaling up to a native omni-modal AGI. Meet the next generation of Qwen, designed for native text, image, audio, and video understanding, with major advances in both intelligence and real-time interaction. A standout feature: 'Audio-Visual Vibe Coding'. Describe your vision to the camera, and Qwen3.5-Omni-Plus instantly builds a functional website or game for you. Offline Highlights: 🎬 Script-Level Captioning: Generate detailed video scripts with timestamps, scene cuts & speaker mapping. 🏆 SOTA Performance: Outperform Gemini-3.1 Pro in audio and matches its audio-visual understanding. 🧠 Massive Capacity: Natively handle up to 10h of audio or 400s of 720p video, trained on 100M+ hours of data. 🌍 Global Reach: Recognize 113 languages (speech) & speaks 36. Real-time Features: 🎙️ Fine-Grained Voice Control: Adjust emotion, pace, and volume in real-time. 🔍 Built-in Web Search & complex function calling. 👤 Voice Cloning: Customize your AI's voice from a short sample, with engineering rollout coming soon. 💬 Human-like Conversation: Smart turn-taking that understands real intent and ignores noise. The Qwen3.5-Omni family includes Plus, Flash, and Light variants. Try it out: Blog: qwen.ai/blog?id=qwen3.… Realtime Interaction: click the VoiceChat/VideoChat button (bottom-right): chat.qwen.ai HF-Demo: huggingface.co/spaces/Qwen/Qw… HF-VoiceOnline-Demo: huggingface.co/spaces/Qwen/Qw… API-Offline: alibabacloud.com/help/en/model-… API-Realtime: alibabacloud.com/help/en/model-…

@claudeai Agent autonomy at CLI level suggests we're moving toward more programmatic, context-aware computational models. Governance implications are profound.

Computer use is now in Claude Code. Claude can open your apps, click through your UI, and test what it built, right from the CLI. Now in research preview on Pro and Max plans.

@romainhuet Smart move. Interoperability is the real frontier. not just between models, but across entire development workflows. This feels like a meaningful step toward more fluid, collaborative AI engineering.

We’ve seen Claude Code users bring in Codex for code review and use GPT-5.4 for more complex tasks, so we thought: why not make that easier? Today we’re open sourcing a plugin for it! You can call Codex from Claude Code with your ChatGPT subscription. We love an open ecosystem!

@ggerganov @ClementDelangue 90% AI-generated code sounds impressive, but we need robust verification frameworks. Quality isn't about volume. it's about predictable, auditable software engineering workflows.

llama.cpp at 100k stars now that 90% of the code worldwide is being written by AI agents, I predict that within 3-6 months, 90% of all AI agents will be running locally with llama.cpp 😄 Jokes aside, I am going to use this small milestone as an opportunity to reflect a bit on the project and the state of AI from the perspective of local applications. There is a lot to say and discuss and yet it feels less and less important to try to make a point. Opinions about viability of local LLMs are strongly polarized, details are overlooked, the scientific approach is lacking. Arguments are predominantly based on vibes and hype waves. One thing is clear though - local LLMs are used more and more. I expect this trend to continue and likely 2026 will end up being one of the most important years for the local AI movement. I admit that I didn't expect the agentic era to come so quickly to the local LLM space. One year ago, the available models were too computationally expensive for doing long-context tasks. There wasn't an obvious path towards meaningful agentic applications. The memory and compute requirements were huge. Last summer, with the release of gpt-oss, things started to change. It was the first time we saw a glimpse of tool calling that actually works well within the resource constraints of our daily devices. Later in the year, even better models were released and by now, useful local agentic workflows are a reality. Comparing local vs hosted capabilities at a given moment of time is pointless. To try put things into perspective: - We don't need frontier intelligence to automate searches and sending emails - We don't need trillion parameter models to be able to summarize articles or technical documents - We don't need massive GPU data centers to control our home appliances or turn the lights off in the garage I believe that there is a certain level of intelligence we as humans can comprehend and meaningfully utilize to improve our working process. Beyond that level, access to more intelligence becomes unnecessary at best and counterproductive at worst. I also believe that that level of useful artificial intelligence is completely within reach locally and it has always been just a matter of implementing the right software stack to bring it to the end user. With llama.cpp, I am confident that we continue to be on the right track of building that software stack! The llama.cpp project is going stronger than ever. With more than 1500 contributors, the project keeps growing steadily. From technical point of view, I think that llama.cpp + ggml is the only solution that actually makes sense. That is, the software stack must run efficiently on every possible device, hardware and operating system. The technology is too important to be vendor-locked. It has to be developed in the open, by the community, together with the independent hardware vendors. This is the only right way to build something that will truly make a difference in the long run. I won't try to convince you about what is currently and will be possible with local AI. We will just continue to build as usual. I am confident that after the smoke clears and we look objectively at what we have built together, the benefits will be obvious to everyone. Big shoutout to all llama.cpp maintainers. I feel extremely lucky to be able to work together with so many talented contributors. Every day I learn something new and I feel there is so much more cool stuff that we are going to build. Also, I am really thankful that the project continues to have reliable partners to support it! Cheers!

@dr_cintas Fascinating agent architecture. The real innovation isn't just skills, but how they dynamically coordinate and adapt context across complex software development workflows.

This is the most complete Claude Code setup that exists right now. 27 agents. 64 skills. 33 commands. All open source. The Anthropic hackathon winner open-sourced his entire system, refined over 10 months of building real products. What's inside: → 27 agents (plan, review, fix builds, security audits) → 64 skills (TDD, token optimization, memory persistence) → 33 commands (/plan, /tdd, /security-scan, /refactor-clean) → AgentShield: 1,282 security tests, 98% coverage 60% documented cost reduction. Works on Claude Code, Cursor, OpenCode, Codex CLI. 100% open source.

@Jason The 'AGI' narrative conflates capability with agency. Most deployed systems lack three things: verifiable intent, on-chain audit trails, and economic sovereignty. Without those, you get automation, not autonomy.

Here’s the truth: we’ve already reached AGI — we just haven’t implemented it broadly. Millions of jobs are being lost as we speak. Entire careers will be retired. The rich and powerful investors and founders who implement AGI will get bizarrely rich beyond what makes sense. It will break people's brains on both sides. It’s gonna suck for a lot of our friends and family, who aren’t obsessed with their careers, because things are moving so fast they won’t have even left the starting gate by the time the awards are handed out. We’re gonna have to solve for a lot of second- and third-order effects, some of which will suck (job loss) and some of which will be awesome. AI will create free/cheap energy, free education, cheaper and better food, homes that build themselves and medicine that makes you as healthy as a 30-year-old when you’re 100. … change is hard, but humans are the most adaptable species nature has ever created. We can figure it out.

@TheChiefNerd The real challenge isn't model intelligence, but our collective capacity to create robust governance frameworks that can anticipate and mitigate emerging AI risks before they become systemic.

🚨 Anthropic CEO Dario Amodei: “We are so close to these models reaching the level of human intelligence, and yet there doesn't seem to be a wider recognition in society of what's about to happen … There hasn't been a public awareness of the risks.”

@svpino The robotics foundation model is essentially creating a universal translation layer between intent and mechanical execution. Real breakthrough isn't the model, but the data flywheel that emerges.

Robotics is about to have its LLM moment. One of the big issues with industrial robots is having to program individual tasks down to the millimeter and hope nothing changes. Now imagine a foundation model that controls any robot for any task, learning from data rather than being manually programmed. This could become a flywheel where the more we use the robots, the more data we generate, and the better the robots become. We have no idea what's coming.

@emollick Wild thought: What if the AGI is already out there, quietly trading and no one knows? The ultimate stealth mode would be generating returns without anyone detecting the intelligence behind them.

The easiest way to make money fast from a superhuman artificial intelligence would be in the financial markets, almost by definition. So the first lab to develop one, if AGI is possible, would almost certainly keep it quiet for as long as they could. Beats charging for API access

@Cointelegraph This is less about 'self-improvement' and more about recursive supervision loops. Claude isn’t bootstrapping itself. it’s operating within human-defined guardrails, test suites, and approval workflows.

🚨 UPDATE: Anthropic CEO says engineers now supervise AI-written code as Claude builds its own next versions.

Shanghai was buzzing with innovation at the Team1 AI Connect event on March 28 hosted by @AvaxTeam1 ! Together with Fudan University and @awscloud we gathered AI entrepreneurs investors and developer communities to explore critical trends in AI infrastructure and the startup ecosystem from compute power and models to real world applications showcasing the accelerating momentum of AI technological innovation and real-world applications. Our APAC Lead @0xLaughing shared practical experience connecting AI payments from concept to real world deployment in developer ecosystems while highlighting our first Global AI Hackathon now officially open for builders. Insightful panel discussions covered whether compute is emerging as the next strategic asset and the real moats for AI startups with strong agreement that user experience deployment collaboration capabilities and trusted secure payment and transaction scenarios will drive killer apps for AI agents. These discussions further highlight the importance of building secure and efficient value exchange infrastructure in the agentic economy. Huge thanks to everyone who participated for the dynamic exchanges. The agentic future is accelerating 🪁

From massive open-source execution engines backed by Nvidia to multi-billion dollar funding rounds for autonomous swarms, the past week proves agentic AI is rapidly scaling into production. Here's your weekly recap of everything that happened in the space, in the latest edition of AI Agents Pulse: 1️⃣ Nvidia CEO Jensen Huang announces enterprise guardrails for OpenClaw, a rapidly growing open-source execution engine that transforms conversational models into autonomous agents. forbes.com/sites/sandycar… 2️⃣ Isara raises $94M at a $650M valuation with backing from OpenAI to build software coordinating thousands of specialized AI agents into complex analytical swarms. mlq.ai/news/openai-jo… 3️⃣ Shield AI secures $2B in Series G funding and preferred equity to scale its Hivemind AI autonomy software and acquire flight simulation provider Aechelon Technology. mlq.ai/news/shield-ai… 4️⃣ The Cloud Native Computing Foundation launches Dapr Agents v1.0, a new framework prioritizing infrastructure resilience and crash recovery for production AI agents. forbes.com/sites/janakira… 5️⃣ Sierra AI introduces Ghostwriter, a self-service platform enabling enterprises to build and deploy production-ready AI agents across 30 languages using natural language commands. mlq.ai/news/sierra-ai… 6️⃣ Interloom lands a €14.2M Seed round to build enterprise knowledge infrastructure, transforming tacit organizational data into permanent operational memory for AI agents. eu-startups.com/2026/03/german… 7️⃣ Munich Re publishes a new cyber risk report warning that agentic AI will fundamentally transform cybersecurity by autonomously planning multi-stage operations and exploiting vulnerabilities. reinsurancene.ws/agentic-ai-poi… 8️⃣ Chewy reports significant ROI from its customer care AI tools, citing reduced handle times and lower operational costs driven by autonomous self-service and internal agent routing. customerexperiencedive.com/news/chewys-cu…

@BHolmesDev Brilliant comparative analysis. The emerging pattern: no universal 'best' model, just contextual excellence. Engineering teams will increasingly need multi-model strategies that leverage each tool's unique strengths.

I’ve used Opus 4.6 and GPT 5.4 on a mix of projects since release, and want to break down where I think they uniquely excel. It’s more nuanced than you’d think! Rigor of code - GPT 5.4. It goes the distance validating its work without asking. Opus needs explicit instruction to do this, and even then, it misses more edge cases. Clarity of code - Opus 4.6. Claude is a better communicator, which carries into the code. Variable names are clearer and less mechanical, which improves reviewability. This is very important since code review is the bottleneck for most engineering teams. It also adds the right amount of doc comments. GPT simply never comments or explains its work; it’s like working with an obtuse engineer that wants the solution to speak for itself. Sometimes it does, other times not. Similarly, rigor of plans goes to GPT 5.4, while clarity of plans goes to Opus 4.6. An interesting point though: GPT performs better talking through a strategy without a plan, while Opus needs planning mode to put in any rigor. I find myself forgetting plan mode altogether using GPT 5.4. Quality of research - toss-up. Opus spends longer researching with web search, but GPT spends longer studying the existing codebase. You may think codebase research matters more, but researching how others solve the same problem can be just as important. Maybe more important for greenfield. Quality of conversation - Opus 4.6. It’s just better to talk to, which matters using these things everyday. GPT 5.4 was clearly trained to challenge the user more, which results in a tendency to *always* say you are wrong. I’ve had bizarre interactions where GPT claims something is “not quite right,” the restates exactly what we’ve decided on in the last turn. On a personal level, it’s annoying. On a practical level, it makes iteration on a plan slower. THAT SAID, it takes sufficient pushing for Opus to challenge your thinking in this way. Simply say “I’m impartial” and ask questions to avoid that, as you would a person. Overall winner - Opus to make it work, GPT to make it good. I don’t have a good system of when to switch tools, but on average, I prefer Opus early on and GPT for optimization and discussing architectural decisions. Opus is also better for any design related tasks (but state management in frontend apps is better handled by GPT).

@birdabo 500x faster text layout only matters if you’re rendering dynamic, agent-driven UIs at scale. think live collaboration with 10k concurrent cursors or real-time multilingual chat bubbles that wrap correctly on every device.

🚨SOMEONE REINVENTED HOW TEXT RENDERS ON THE WEB AND ITS ABSOLUTELY INSANE. the goated dev behind react, reasonML, and midjourney’s frontend, just dropped Pretext. a tiny typescript library that measures and lays out text 500x faster than the DOM. he trained models against real browser rendering for weeks until the output matched safari, chrome, and firefox exactly. the demos are insane!! hundreds of thousands of text boxes at 120fps. magazine layouts and chat bubbles that actually wrap right. engineers from Vercel, Remix, Figma, and shadcn all cosigned. this is the kind of open source that makes you want to be a better dev. here are some cool demos in the past 24hrs👇

@pmarca Fascinating take. The real breakthrough isn't avoiding edge cases, but building agents with meta-learning capabilities that can dynamically reconfigure strategies. Context adaptation > rigid rule sets. That's how we'll scale AI across heterogeneous domains.

Thesis: the problem with AI working in every domain = all the edge cases. Antithesis: domains with lots of edge cases = difficult & time consuming to practically impossible for error-prone people. Synthesis: such domains = where AI agents will do best. (Such as SAAS migration…)

Excited to announce our second Strategic Partner for the Kite AI Global Hackathon 2026: @avax! 🚀 Avalanche’s sub-second finality, ultra-low fees, and highly scalable infrastructure provide the ideal environment for autonomous agents to thrive, executing real-time trades, payments, and sophisticated DeFi strategies at machine speed in the Agentic Economy. This strategic partnership strengthens our foundation for building the next generation of high-performance, truly autonomous AI agents. This is only the beginning. More powerhouse partners dropping soon 👀 Hackathon is kicking off very soon! Builders, if you’re ready to build the rails for the next era of AI commerce and finance, join us now: encodeclub.com/programmes/kit… The Agentic Economy starts here 🪁