Mohan's Google chapter: Antigravity inverts developer acceleration. Not faster writing but parallel orchestration. Sergey Brin ships PRs through it. linkedin.com/pulse/weightless-architect-varun-mohans-google-chapter-robert-schwentker-yhtlc

Ilya Sutskever just confirmed the "Age of Scaling" is over. The next AI moat isn't compute; it's "taste." My analysis of his rare new interview on the shift to the "Age of Research," the "Value Function" & what it means for enterprise strategy. linkedin.com/pulse/age-re...

Klimt's Portrait of Elisabeth Lederer sold tonight for $236.4M--2nd highest auction price ever & 100% jump over his previous record. Painted 1914-16, hung above Leonard Lauder's dining table for decades. 20-min bidding war. The room erupted! 🔔

MCP's new Tasks primitive (SEP 1686) shifts agent architecture from orchestration to choreography. Clients manage task lifecycles instead of blocking on synchronous calls. linkedin.com/pulse/async-threshold-mcps-infrastructure-moment-arrives-robert-schwentker-wasgc/

Curious that what looked like fragmentation now reads as specialization Google optimizes tempo Cognition optimizes insight Agent RFT shows insight can train itself. linkedin.com/pulse/learni...

At yesterday's Web Discovery Summit 2025, @paraga @DhruvBatra_ @sarahmsachs explored a new web built for patient machines that read, negotiate, and never sleep. The question: When agents become users, what still belongs to humans? www.linkedin.com/pulse/agent-...

If relevance now decays in minutes, what replaces attention as the web’s true currency? The answer, argues "Half-Life Problem" is continuity --data that regenerates instead of rots. linkedin.com/pulse/half-l...

If AI agents can buy, who do they trust? If tokens self-destruct, what anchors value? Agentic Commerce Protocol (ACP) turns payment from final click to ongoing negotiation. Secure Payment Tokens (SPT): sealed letters of credit that expire. www.linkedin.com/pulse/when-a...

“Memory is more than context windows” — @charlespacker on stateful agents & long-horizon reasoning. Check out my recap of the Intuit meetup: “Stateful Futures: @Letta_AI, @Intuit & Architecture of Perpetual Agents” → linkedin.com/pulse/statef... #AI #agents #infrastructure

Tonight in San Francisco QR-coded greenroom door @Vapi_AI set the tone: automation w/o awareness. Agentic = behavior, not chrome. Generation outruns comprehension. Execution automates, taste does not. w/ @juliusai @magicpatterns @withdelphi www.linkedin.com/pulse/design...

.@beyarkay Boyd Kane’s essay on why AI can’t be “debugged” like software is lighting up @hackernews - engineers & executives suddenly realizing they’ve been seeing the same system through incompatible lenses. The overlooked risk in AI deployment ↓ linkedin.com/feed/update/...

When agents spend on behalf of humans, where does intent reside? Stripe & OpenAI just architected the payment rail for AI commerce. But the protocol reveals something deeper: we're building infrastructure faster than we understand its implications. linkedin.com/pulse/protoc...

OpenAI DevDay 2025 shifted the conversation from "do agents work?" to "who controls the infrastructure they run on?" New analysis on AgentKit vs LangGraph, CrewAI, LlamaIndex & n8n - comparing not just features but philosophies about where complexity should live. www.linkedin.com/pulse/openai...

What processes assume software takes weeks to build? If intelligence compounds monthly what does annual planning mean? When software gens itself what becomes the product? 3 questions as autonomous agents build Slack in 30hrs & replicate APIs in 5hrs www.linkedin.com/pulse/softwa...

380 AI events in 7 days reveals where the industry actually moved. From "what if" to "how to." From demos to deployment. Agents, MCP, infra & policy dominate. The calendar doesn't lie. #SFTechWeek www.linkedin.com/pulse/380-ai...

Alex.com: Applicant volume tripled. Time to hire hit 60 days. "You can close a house in SF faster than hire an engineer." The AI recruiting conundrum: using machines to help humans get jobs, not take them. New analysis on what changes in the next 12 mos: linkedin.com/pulse/ai-rec...

Model Context Protocol (MCP) update: * Asynchronous long-running operations * MCP Registry API * Server IDs * Official extensions * SDK tiering ...all land Nov 2025 - marking MCP’s shift from experiments to backbone for AI integrations. My thoughts: linkedin.com/posts/schwen...

Last night @ Frontier Tower: 10 builders showed invisible infra being laid beneath AI agents. Less "look what AI can do" - more "here's scaffolding we're quietly building." Commerce layers, auth protocols, data normalizers. www.linkedin.com/pulse/mcp-de...

.@DrHowardGardner & @AntheaERoberts rethink knowledge work in the age of AI. From dragonfly vision to the second Socratic inversion, the future may belong to question-architects, not answer-providers. my thoughts: linkedin.com/pulse/when-c...

From click-to-buy to cryptographic consent. Google’s new AP2 protocol could be the missing trust layer for agent-led commerce. My analysis here: 🔗 www.linkedin.com/pulse/from-c...

. @levie.bsky.social "Imagine the 50 person company where all of a sudden they can act like a 500 person company" ...on AI's leverage effect. Small teams, big reach changes everything. www.linkedin.com/pulse/ai-mom...

Agents can unlock the 90% of work stuck in unstructured content, but only if we redesign how we work. Agent-first design, context engineering, platform literacy. #BoxWorks 2025 San Francisco @levie Keynote My recap: www.linkedin.com/pulse/strate...

BoxWorks VP panel exposed the AI velocity paradox: 🚀 Releases faster than ever 🏢 Enterprises can’t absorb change Insights: capability frameworks > feature training, platform literacy > silo adoption. Part 3 of 4 in my BoxWorks 2025 series: www.linkedin.com/pulse/execut...

Conversations from the trenches of enterprise AI: * Integrators debugging MCP servers. * Banks modeling AI unit costs. * Peace officers trained in 15 min bursts. * Consultants building brand voice agents. The bottleneck isn’t tech...it’s education. www.linkedin.com/pulse/boxwor...

87% of orgs piloted AI agents, only 24% have governance. The gap isn’t tech...it’s education imho. From MCP debugging to 15-min peace officer training, BoxWorks 2025 shows the last mile of enterprise AI. Article 2 of 4 👉 linkedin.com/pulse/boxwor... w/ @vishalmanohar @ShawnDeines

BoxWorks 2025 spotlight: protocols like A2A, ADK & MCP shaping intelligent content: content that reasons, remembers & complies. ⚖️ Which workflows in enterprises demand deterministic sequences for compliance? linkedin.com/pulse/boxwor... @Box @AliArsanjani @shurrey.bsky.social

Defeating Nondeterminism in LLM Inference by Horace He (Thinking Machines' blog, an AI lab founded by Mira Murati) Connectionism will cover topics as varied as their research is: from kernel numerics to prompt engineering. Here, they share what they are working on.

This is how one signals ones building something different The choice to focus on determinism isn't randon: it's a direct line to enterprise-grade AI where reliability isn't a feature, it's a requirement Seems to me they're playing a long game starting w/ trust & repeatability.

Developer education 🧩 is shifting fast My latest dive: New Playbook for AI Dev Ed Prediction🔮 Next 6 months we’ll see rise of voice-powered agent-driven content development: --> new paradigm for how AI 🤖teaches & learns alongside us #AI #DevEd linkedin.com/pulse/new-pl...

Experimenting with @para.ga new Deep Research system @p0 Parallel Web Systems. Can multi agent AI tackle one of the hardest challenges in education - keeping content accurate, adaptive & alive? 🔗 www.linkedin.com/pulse/from-c...

Analysis of UC Berkeley Agentic AI Summit - where enterprise leaders from @glean @SierraPlatform @Get_Writer @Replit revealed what's really happening beyond demos. Reliability gap btw prototype & production is real, but so are the solutions emerging. linkedin.com/pulse/uc-ber...

.@rweiss57 "The next leap in agents isn’t just search - it’s precision context delivery. @tavilyai vision: one line of code to connect agents to web & proprietary data, returning only the most relevant snippets. Less noise, more signal, faster answers. #RedisReleased

Human-like memory for agents means layered, weighted associations, not just short-term vs. long-term buckets. @cognee_ is building modular memory pipelines where frequency, recency, & rules guide retrieval, just like our mental lexicon. #RedisReleased Vasilije Markovic

In large, evolving codebases, context is the bottleneck. @augmentai is pushing coding agents beyond autocomplete, into a shared workspace memory that understands conventions, mistakes, & workflows, enabling automation w/o babysitting. @guygr #RedisReleased

From demos to production, the future of AI agents depends on reliable context & memory. LangGraphAI checkpointing + flexible architecture could make multi-agent systems as dependable as traditional software but w/ adaptability only LLMs bring. @samecrowder #RedisReleased

At #RedisReleased dev conf in SF yesterday, AI startups explored how agents learn, remember & act with meaningful context. My takeaways on tackling the memory challenge: linkedin.com/pulse/memory...

Experimenting 🔧 with the Box MCP Server + Inspector Watching live tools pop up for auth, search, upload, analyze Fascinating to explore how this protocol opens up new ways to interact with content systems Repo: github.com/box-communit...

Hello World! @theworldlabs marble.worldlabs.ai/world/8772db...

Zero to product in a day. Buildathon by AI Fund + DeepLearning.AI: 5 projects, Speed Challenge, live finals. w/ Andrew Ng - takeaways & winners: linkedin.com/pulse/builda... #AI #GenAI #buildathon

Kickoff at @Citi with @BlazeOByrne: Gen AI Impact on Software Development. Thanks @real_ioannis @reflection_ai @jasoncwarner @poolsideai.bsky.social @cramforce @vercel.com @russelljkaplan @cognition for sharing where SDLC is headed. Read thru notes → www.linkedin.com/pulse/decade...

Web shifting from prebuilt code to just-in-time generation. @vercel.com CEO @rauchg.blue: 'Next 3 yrs kingdoms collapse & AI-native companies rise quickly' www.linkedin.com/pulse/genera...

When you use GPT-5, you feel the intelligence shine through… it did in minutes what used to take weeks.” @cursor_ai @mntruell @OpenAIDevs New article: The GPT-5 + Cursor Moment: Strategic Realignment in AI Dev www.linkedin.com/pulse/gpt5-c...

Patience. #gpt5

Dario Amodei just decoded 🧬 learning itself: Children learn📋 two ways: watch👁️ parents OR experiment🧪 w/ the whirled 🌎 AI mirrors🪞 this exactly: Base training (imitation) + RL⚖️ (trial & error). "Chain of thought" is just fancy RL where the model writes, acts, learns ⚡

"Promptware refers to prompts engineered to behave like malware, exploiting the advanced capabilities of LLMs to execute malicious activities." sites.google.com/view/invitat...

"Three weeks. That's how long it took Cognition to transform an acquisition into an ultimatum." The Windsurf paradox reveals how AI consolidation can accidentally optimize away the very intelligence it seeks to acquire. www.linkedin.com/pulse/windsu...

AI reasoning models are reshaping business strategy faster than most leaders realize. New content paradigms emerging around conversational AI will separate winners from laggards in 2025. linkedin.com/pulse/reason...

"Software was always disruptor. Now software itself disrupted. Software is eating software." @martin_casado New deep dive: How AI fundamentally inverts economics, transforms billing into intell & forces cos to rethink everything from dev roles to pricing strategy www.linkedin.com/pulse/ai-eat...

While Windsurf got dismantled in 72 hours, @replit quietly hit $100M ARR with 45% monthly growth. The difference? One bet on model optimization, the other built for model independence 🧵 www.linkedin.com/feed/update/...