Questflow @questflow

Pick Grok 4.7 as the brain behind your AI Finance Agent. Try it now! next.questflow.ai

Grok 4.7 is now live on Questflow 🚀 The latest flagship model from @SpaceXAI, and a notable upgrade over Grok 4.6 at the same price and speed! Now available on Questflow 👇

Curious how we actually use AI inside Questflow? @Bobbxu breaks down where AI helps the team move faster, and where humans still need to make the call. This one’s worth watching 👀

This is a Pod @TiPodcastt

a day ago

AI creates a first draft. People make it better. @Bobbxu and @lp_coder share where AI helps their teams and where human judgment still matters.

0 5 9 540 1

AI creates a first draft. People make it better. @Bobbxu and @lp_coder share where AI helps their teams and where human judgment still matters.

@Tonyyofficial24 @Bobbxu 🫡

If you’ve ever been curious about how Questflow got started, this one’s for you. Our CEO @Bobbxu shares how Questflow evolved from a general AI workflow product into what we’re building today, and some of the lessons we learned along the way. A little bit of Questflow history, straight from the founder.

This is a Pod @TiPodcastt

6 days ago

What does AI mean outside the tech bubble? TiP starts with people, their stories, and everyday life. Every story finds its voice！ Sponsored by @questflow and @PromptiveOne

1 9 15 833 0

Jev from @typesafeai has been everywhere lately. If you had Jev inside your trading workflow, what would you actually hand over to it? Buy or wait? Take profit or let it run? Cut the loss or give it more room? Or maybe just one very useful answer: don’t take this trade. What decision would you trust an AI to make for you?

Meta upgraded its planned data center from 2 gigawatts to 5 gigawatts within a single year, part of a broader AI buildout that's created real demand for the suppliers keeping those clusters connected and cooled, not just the GPUs inside them. Five names sitting in that lane, with real numbers behind them: $COHR (Coherent): FY2026 revenue $7.12B, up 23%. Datacenter and comms segment alone did $5.27B. Q1 FY27 guidance sits at $2.2-2.4B, driven by AI optical engines and lasers. $AAOI (Applied Optoelectronics): 2025 revenue $456M, up 83%. Q2 2026 revenue $192M, up 86% YoY. TTM revenue now $596M. $NBIS (Nebius): AI cloud infrastructure name that's had one of the sharpest re-ratings in the sector this year as hyperscaler compute demand outpaced supply. The demand signal behind all of this: Nvidia reportedly asked optical suppliers to scale indium phosphide laser capacity 20x through 2030 to keep pace with AI cluster networking. Suppliers countered with 12x, worth noting, since that gap is the market's own hedge against overbuilding. The InP optics market itself is projected to grow from $1.9B in 2025 to $22.75B by 2030. The infrastructure thesis here is real and has verifiable revenue growth attached to it. Any specific return claim attached to it, past or projected, is a separate thing entirely and worth your own due diligence before acting on it.

Yeah, we’ve seen the @Muse hype too. Quick reminder: Muse 1.3 is already on Questflow. Pick it as the brain behind your Agent and build your first AI Finance Agent for research, analysis, and trading. If you’re already deep in the Muse rabbit hole, you might as well put it to work.

Questflow @questflow

3 weeks ago

Muse Spark 1.3 is now live on Questflow! A stronger base model means a stronger Finance AI Agent. With longer context, stronger reasoning, and more reliable agentic execution, the Finance AI Agents you build can handle more complex market workflows, process more information,

4 4 21 2K 2

x.com/i/article/2100…

Intel CEO Lip-Bu Tan flagged where AI infrastructure spending shifts next: memory, connectivity, optics, power, and cooling, not raw compute, as the emerging bottlenecks. Five names sit directly in that path: $CRDO (Credo): GPU-to-GPU connectivity as clusters scale, supporting speeds up to 1.6 terabits per second $ALAB (Astera Labs): Switches, retimers, and controllers linking GPUs, CPUs, and memory as computing shifts to full AI racks $CLS (Celestica): High-speed networking hardware moving beyond contract manufacturing into AI infrastructure, its DS6000 platform delivers 1.6T Ethernet connectivity $AAOI (Applied Optoelectronics): Optical transceivers with hyperscaler orders already locked in for both 800G and 1.6T $LITE (Lumentum): Lasers and photonic components as copper hits physical limits on distance, bandwidth, and heat. Nvidia's own $2B investment and multiyear purchase agreement signals how strategic optical capacity has become As GPU count scales, keeping thousands of chips connected and fully utilized becomes its own engineering problem, one that's arguably less crowded than the GPU trade itself.

Anthropic just extended Claude Code from single conversations into persistent, parallel work: describe a project once, and Claude runs multiple threads that keep working even after the laptop closes. That's a meaningful shift in interaction model, from "prompt and wait for a response" to "delegate and check back later," closer to how a team lead assigns and tracks parallel workstreams than how a chat interface has traditionally worked. Currently in beta for select Pro and Max users in cloud sessions, with broader rollout planned. Worth watching how this holds up on longer, more ambiguous tasks, that's typically where autonomous multi-thread execution reveals whether it's genuinely reliable or just impressive in short demos.

Palantir CEO Alex Karp added a third position to the AI safety debate this week: not "regulate it" and not "let the market handle it," but hold builders personally accountable. "The first line of defense is you're liable for your own actions," he told CNBC, arguing that civil and criminal liability, not new government rules, should be the actual check on companies claiming their own technology could be catastrophic. His sharper point: he suggested some AI labs may be pushing for regulation specifically to avoid that liability, "I want new rules for myself, and I don't want to pay the true cost of that." Karp went further than most in the debate too, floating that frontier AI labs may eventually need to be nationalized, arguing that without state involvement, "every single one of my clients is going to sue" over stolen business data absorbed into closed models. Worth watching whether liability exposure or new regulation ends up being the actual lever that changes lab behavior, Karp's betting on the former.

Former SEC Chair Gary Gensler told CNBC the AI trade is "unlikely to last," warning today's infrastructure buildout will eventually plateau and reverse, "it's just a question of when and how steep." His mechanism is specific: if AI companies stop beating quarterly earnings estimates and start flatlining, "the capital markets will reprice it," starting with memory chip suppliers losing pricing power first. Worth noting the timing lines up with real market movement, the PHLX Semiconductor Index dropped over 5% in the five days following Amodei's and Altman's calls for an AI slowdown, and AI capex is projected to exceed $795B in 2026, heading toward $1T in 2027. Gensler flagged private credit, vendor financing, and loan guarantees tied to that buildout as the areas to watch if the cycle actually turns, that's where the real fragility would show up first, not in any single company's stock price.

@garyvee Good read, most "big" trends look small right up until they don't. If you've got something worth selling, now's the cheap-experiment window before it gets crowded.

@AltcoinDaily $250k is the headline that gets clicks, the Fed/CFTC/tax bill combo is the part actually worth watching closely. 👀

@elonmusk @bot Voice mode changes how people actually use these things day to day, curious how it holds up outside a demo.

@scottmelker Fair bet, moving fast without rules written tends to make regulators regret the delay more than the industry. 😄

Stop using AI like a chatbot. Start working with it like a teammate. Research together. Think together. Trade together. Meet your Questflow AI Finance Agent 👇 next.questflow.ai/trade

We gave your AI Finance Agent a seat at the trading desk. Questflow AI can now follow what you see, understand what you click, and stay in context as you move through the market. Open a chart. Explore a trader. Check a signal. Switch markets. Your Agent stays with you through it all. Research with you. Think with you. Trade with you. Your AI Finance Agent is no longer just a chat. It’s your teammate on the desk.