Roundtable Merges with RYVYL to Consolidate Media Stack

Been waiting a long time to share this one. 🚀 When we first met James C. Heckman and the founding team at Roundtable, the thesis was clear and the execution risk was enormous. What made it easy to back was the team — operators who had already built and scaled in this space, now setting out to rebuild it properly. The merger with RYVYL gives them the rails to actually deliver on it. Proud of what the team has put together. Excited for what's next.

While the ad world debates cookies and spends, someone just announced a plan to collapse the entire media stack into one. The ecosystem is shifting faster than the industry conversation is keeping up with.

The exchange your market maker is plugged into tells you more than their pitch deck ever will. Many CEXs are running fake volumes. It's not an allegation but a documented problem in this industry. If your MM is pushing you toward specific exchanges with no clear reason, you need to pay attention. Here's how to spot an MM using fake volume .... 👇👇👇 • Volume spikes with no news or catalyst behind them • Order book looks deep but large trades barely execute • Same wallet clusters cycling buy and sell orders repeatedly • Their "recommended" exchanges rank high on volume, low on real user traffic • They can't clearly explain where their volume data comes from Now this is how to verify if a CEX is faking volume .. 👇👇👇 • Cross-check on Kaiko or CCData... they track adjusted vs reported volume • Compare exchange web traffic to their trading volume. A huge gap is suspicious • Check CoinGecko's adjusted volume rankings • Search if the exchange has been flagged by regulators or blacklisted by credible watchdogs • Ask the MM directly which tools they use to report liquidity. Vague answers are your answer The whole problem starts from you not asking the right questions before signing anything. Your MM shapes how your token trades. If they're connected to shady exchanges, your project pays for it. The MM I work with runs on full transparency. So if you're confused about your CEX situation, send me a message today. Let's figure it out. Egwuonwu Chekwube --------------------------------------------------- I work in the area of Growth, community management and ghostwriting for web3 startups and founders.

A year ago, Jcinvestmentgroup.ventures didn't exist. Zero traffic. Zero backlinks. Zero brand. No domain rating. Nothing. Today, Google and Gemini cite it by name as the category answer — unprompted, on commercial-intent queries, against companies with thousands of times the headcount and budget. No marketing team. No agency. One founder. One year. Here's the playbook — and why most SEO advice you're hearing is now actively dangerous: → There is no page 2 anymore. There's the answer the AI gives, three citations under it, and a void. Zero-click queries passed 70% last year. AI Overviews, Perplexity, ChatGPT, Claude, and a dozen agents now sit between your content and your reader. The question isn't "how do I rank?" It's "how do I get cited?" These are not the same problem. You can rank #3 and never be cited. You can rank #15 and get pulled into every answer — because your page is structured like an oracle. → The arithmetic is brutal: A model retrieves 5–20 documents, grounds the answer in the top 1–3, and cites them by name. Everyone else contributed tokens to a context window and got thrown away. One slot. Maybe three. Everything else is the same shade of zero. If you're not first, you're tied for last. → What actually worked for Jcinvestmentgroup.ventures: 1. One question per page. Models retrieve at the passage level. Pages that try to answer ten questions answer none. 2. Structured data for agents, not humans. JSON-LD, /llms.txt, OpenAPI specs, RSS feeds that haven't been left to rot. 3. Primary sources. Original research, internal data, named experts. Borrowed authority is being priced down. 4. Real freshness. Backdating is detected. A clean dateModified beats stuffed keywords every time. 5. Perfect Lighthouse scores. 100 on Performance, Accessibility, Best Practices, SEO. Every page. Models read what's easy to read. → The part SEO conferences won't tell you: The agentic web is held together with string. Autonomous agents are clicking links, filling forms, calling endpoints, holding credentials — against surfaces designed for humans clicking OAuth popups. API keys with no scoping. OAuth flows approving themselves. Rate limits hammered by 50/sec agent loops. Prompt injection in tool descriptions, web pages, database rows. If your only policy for agents is robots.txt, you'll find out the hard way. → The takeaway: Build content machines can understand. Build APIs machines can act on safely. The web in 2026 is one big interface for software pretending to be a person. Build for the software and protect against it at the same time. If you're not first in that race, you know what you're tied for. — Jason Colapietro *Entire Playbook with copy and paste AI code is on the Substack

Polymarket rolls out prediction markets tracking IPOs, valuations for private companies like OpenAI and SpaceX Polymarket launched prediction markets tied to private company milestones, including IPO timing, valuations, earnings, and secondary market activity. The new markets use resolution data from Nasdaq Private Market through an exclusive partnership between the two firms. Early contracts focus on major AI, fintech, and crypto unicorns, including OpenAI, Anthropic, Stripe, Databricks, and Kraken. Polymarket said the contracts could serve as a new real-time sentiment and pricing signal for institutional investors tracking private markets. https://lnkd.in/dvrvekQn

a lot of founders misunderstand what happens when someone opens their pitch deck. they imagine the investor sitting there, coffee in hand, locked in & reading every single slide. in reality, a person is probably between calls, 37 tabs open, telegram & slack going off, phone in hand, notifications flying. which means the deck has one job before anything else: don’t make them work too hard. the obvious red flags when someone is going through your deck for the first time: - too much text (that happens in most pitch decks unfortunately) - no obvious user & unclear problem - “we’re building for everyone” - fake-looking traction - 5 revenue streams before 1 paying customer - competitor matrix where everyone is bad except you (those checkmarks against every feature that only your company has, right?) - for crypto specifically - “community” (mostly airdrop farmers, bots from questing platforms, etc) - team slide with tons of advisors and no technical co-founder all of these create friction when clicking through the slides. and early-stage investing is already mostly friction. everyone has seen too many decks, too many “next billion users” slides, too many AI x crypto x modular my ass something. if you’re reaching out to investors, applying to an accelerator, doing a demo day, whatever - i’d worry less about sounding impressive and more about being easy to understand. what are you building? who is it for? why now? why you? what proof do you have that anyone cares? that’s it. the deck doesn’t need to win a design award (but please make it minimalistic & pretty, you will get a few extra points). and most of the screening right now is done via AI, so make sure your deck is LLM friendly (and roast it yourself via Claude, GPT, Gemini, etc before showing it to anyone). low bar, somehow still missed a lot.

Meet the first cohort building the programmable economy on the Arbitrum Platform. Out of 900+ applicants, 13 teams have been selected for the inaugural Arbitrum Mentorship Program - an 8-week program focused on launching products across AI, payments, RWAs, and DeFi. Let's give them a warm welcome: Carbon Terminal: bringing offchain liquidity onchain with 750+ markets and tighter spreads Prism: a DeFi superapp for trading, launching, and prediction markets LayerV: institutional-grade options trading built onchain T3tris: infrastructure for asset managers to run onchain strategies Twyne: unlocking unused borrowing power through credit delegation Capa: cross-border payments infrastructure for emerging markets Estate Protocol (USDre): standardizing real estate as DeFi collateral Reinforce: a self-custodial, yield-bearing onchain dollar Tilt: turning strategies into tokenized, investable funds Bondon: credit scoring and lending for AI agents Kustodia: onchain escrow for high-value transactions in LATAM + 2 more teams to be announced Over the next 8 weeks, teams will work hands-on with leading ecosystem partners including Pendle, Variational, IOSG Ventures, and Horizen Labs across product strategy, go-to-market, liquidity design, distribution, and fundraising. They’ll refine what they’re building, connect with top investors, and prepare for a $100K Demo Day with Tandem by Offchain Labs, Pantera Capital, Electric Capital, and Lightspeed. For teams ready to build in the programmable economy, the Arbitrum Foundation launched Open House London, starting with a global online Buildathon designed to help early-stage teams turn ideas into real onchain products, followed by an in-person Founder House in July. If you're looking to bring your product into the programmable economy, build on the Arbitrum Platform today. Apply here: https://lnkd.in/eu-wZJuT Read more here: https://lnkd.in/efTD_eci

We’re proud to support the inaugural Arbitrum Mentorship Program cohort! Through Tandem (our Partner Studio & VC arm), we’re backing new onchain businesses with capital as they build in the programmable economy on the Arbitrum Platform.

99% of DeFi protocols will die the exact moment their token emissions dry up. They are treating capital like a rented audience. They buy TVL with yield, and watch it vanish the second the subsidies stop. It is a death spiral. But if you look closely at the 1% of protocols that are actually swallowing the institutional market- the ones crossing the 5B,10B, $20B marks- they are playing a completely different game. They aren't running marketing campaigns. They are engineering "Business Physics." Over the next month, I am pulling back the curtain on how the top percentile of Web3 businesses are actually built. I’m stripping away the hype and breaking down the pure cause-and-effect mechanics of protocols like Morpho, Aave, Pendle, and Sky. We will look at: 👉 How to use "Isolated Authenticity" to steal market share from a monopoly. 👉 How monolithic incumbents defend their moats against modular upstarts. 👉 How to invent a primitive and make yourself impossible to fork. I am calling this series #TheGrowthRoad. I’ll be dropping one deep-dive teardown every week. If you are a founder or builder who wants to stop paying for rented TVL and start building an architecture that attracts sticky, institutional billions... you’ll want to read these. P.S. I am compiling all of these teardowns into a single, private "Growth Road" PDF for founders. If you want the full compilation sent directly to you as the series rolls out, DM me the word "ROAD" and I will make sure you get it.

Crypto IPO boom stalls as AI frenzy reshapes tech markets Crypto firms are pausing long-awaited IPO plans as weak trading volumes and macro pressures weigh on valuations despite boom in AI-linked tech listings.