Lorenzo Belli @BlockLollo

2. Inside HQ, Codex and other AI startups have dedicated offices where they hold office hours for NVIDIA employees. Never seen anything like this! I assume because no other large tech company is the kind of "platform" for so much of the tech industry like NVIDIA. The Codex team is v popular here btw as I heard!

Structural reasons to hold $MSTR 1. @Strategy is a corporation embedded in market cap weighted index funds. As balance sheet value and market cap rise (price + share issuance), Strategy’s weight in those indexes increases, forcing passive funds to rebalance and buy more. An estimated 50–60% of net new equity capital flows into passive index vehicles. This is reflexivity (cuts both ways) 2. Leveraged Bitcoin exposure with effectively zero explicit expense ratio versus spot BTC ETFs. 3. Deep, liquid options market on top of liquid spot, enables sophisticated hedging and yield strategies. 4. Bitcoin is used as productive capital, taking calculated risk to generate BTC yield instead of sitting idle. (Increasing “potency” of Bitcoin exposure per share) 5. The $STRC digital credit “machine” targets the largest TAM on the planet: global credit. 6. The “replacement cost” of rebuilding the MSTR BTC treasury (4% of network), capital markets team, AI operating business, and index inclusion, is plausibly in the $75–100B range. 7. Convexity: structurally higher beta than BTC, with upside that can outpace the underlying asset over a full halving and business cycle 8. Transparent, frequent capital markets communication, ability to price risk 24/7/365 9. Can be held in IRAs, 401(k)s, HSAs, and standard brokerage accounts where pure BTC exposure is still constrained. 10. Ability to build a holistic Long BTC portfolio across BTC correlated risk/return profiles (STRC / BTC / MSTR). Structural reasons to trade $MSTR 1. Reflexivity of passive, market cap weighted index flows cuts both ways; forced buying on the way up, forced selling on the way down. 2. Deep options and spot liquidity support tactical positioning, volatility trading, and basis trades. 3. Clean ability to hedge BTC exposure in both directions with structural leverage through the equity and derivatives. 4. Speculation surrounding sentiment, funding, and non‑consensus data points (mNAV, index inclusion/exclusion risk, STRC pricing, issuance cadence, etc.). How I frame the position - Just because you don’t want to hold $MSTR for X/Y/Z reasons doesn’t mean the rest of the market agrees. Roughly 60% of the equity market operates like a rules based “zombie” that buys market cap weighted index funds; those rules, not emotions, drive flows. - Actively holding spot $MSTR means you are structurally long a Bitcoin bull thesis. It will trade with more volatility than BTC; historically, $MSTR has run at roughly a 1.5x rolling beta to Bitcoin. - If your mental model is a 4-year holding period for BTC (one business cycle/halving), the mathematically appropriate duration for $MSTR is longer given the higher beta and volatility. There will be periods, (like right now, and like 2022) when the market (X especially) is extremely noisy and misses the forest for the trees. If you liked the $MSTR thesis at $500, you’re now being offered the same balance sheet and strategy at a ~80% discount. I’m still incredibly long $MSTR, with a very meaningful share of my net worth in the equity. Do I enjoy being down ~80% from ATH? Of course not. But after 12 years in the capital markets, I’ve learned that REAL capital is brutally hard to accumulate. I’ve watched insurance companies grind for years to make ~$60M on a $1B balance sheet, with massive overhead, heavy regulatory drag, underwriting risk, and aging legacy tech. By comparison, a Digital Credit capital structure like Strategy’s is the most efficient and antifragile capital accumulation machine I’ve ever seen (Name another corporation that has gone from $250M to $60B in capital in under 6 years). That’s why, despite the drawdown, I still want exposure to this vehicle on top of my BTC. I trust the capital structure, I trust the team.

David Hoffman @TrustlessState

5 days ago

Why would anyone hold MSTR? @PunterJeff on our pod, you called this 'amplified BTC exposure' Does Saylor minting new MSTR go against this concept? Seems more like to me that MSTR is new fiat currency that Saylor prints from

48 9 101 96K 14

@GergelyOrosz They also changed the performance review process company wide to be more AI focused and leaner

Full scenario: europe2031.ai Executive summary: europe2031.ai/summary/ Pdf: europe2031.ai/europe-2031.pdf Audiobook: open.spotify.com/episode/0c8iIv…

GPT-5.5 lies constantly, Grok 4.20 doesn’t. We made a simulation to see if AI lies when the stakes are life & death.🧵

JUST IN: 🇯🇵 Japan advances bill to reclassify cryptocurrencies as financial products and cut taxes from 55% to 20%.

Found the Best Resource to learn Harness Engineering. 😭 walkinglabs.github.io/learn-harness-…

Rare to hear a founder of an AI coding harness be this honest. A reason I respect @thdxr "We’re shipping way more hacks where we should have just rethought the whole system from the ground up, or redesigned it to make it more flexible. So I think our judgement is just off.”

The Pragmatic Engineer @Pragmatic_Eng

3 weeks ago

AI agents keep placing “landmines” in the code. Dax Raad(@thdxr), creator of OpenCode, on why the human feedback loop kept code quality high: Dax: "We're shipping way more hacks in places where we should have just rethought the whole system from ground up or redesigned it and

6 12 168 162K 79

Read the full report for free here. delphi.link/Tokenmarkets

m e t a d a o

Delphi Digital @Delphi_Digital

4 weeks ago

Our new report "State of Token Markets" is live! Most tokens spend their lifetime below launch price. Across 540+ tokens launched since 2020, the average token spent 70% of its life below launch. Tokens launched at inflated FDVs with minimal float, which handed the initial pop

23 26 146 70K 65

Just incredible interest in @MikeFritzell's piece on how to use Claude for Equity Research and Investing. He goes into quite a lot of depth. 🎓

Anthropic engineer: "You're not supposed to prompt Claude. You're supposed to build a system that prompts itself." this is one of the best workflows I've seen in a long time in this video she breaks down exactly how most people are using Claude: - the 14% you lose to CLAUDE.md before typing a word - the automation workflows most users don't know exist - the daily task pipelines that run without touching the keyboard - the daily workflows Anthropic's own engineers automated first if you've been using Claude for more than a month and never left the chat window, you've been using one agent when you could be running a team of them instead of another show tonight, watch this make sure to bookmark it before it gets lost in your feed the guide is in the article below

Khairallah AL-Awady @eng_khairallah1

4 weeks ago

x.com/i/article/2061…

35 78 488 1.2M 2K

Anthropic engineer: "You can build 5 assistants in one afternoon. Each one handles a task you've been doing manually every single day." In 45 minutes he builds 5 focused agents from scratch on camera. Most people are still doing code review, testing, and documentation by hand every single day Watch the session, then save all templates below 👇

rody @0x_rody

4 weeks ago

x.com/i/article/2061…

18 49 411 1.1M 2K

introducing vyupgrade it automatically rewrites old vyper contracts into modern vyper, then proves the rewrite is safe by checking it compiles under both source and target versions, diffing the abi, method ids and storage layout. it supports all syntax changes from 0.2.1 to 0.4.3, as well as picking up dependencies and modules like snekmate. some rewrites are not inherently safe and need your judgement. the tool flags those and shows you exactly what to look at. $ uvx vyupgrade contracts/ github.com/banteg/vyupgra…

We used these lessons to enforce a much tighter review process for our own tasks. Dive deeper in our blog: deepswe.datacurve.ai/blog Check out the dataset on GitHub: github.com/datacurve-ai/d…

Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks. On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.

Introducing: Browser Use Terminal. A Rust harness + TUI that gets real work done in the browser.🦉 Browser Harness gave LLMs freedom in Chrome. We built a full LLM harness around it - in Rust. > Direct CDP — raw browser control > Real Chrome — use your logged-in browser > Rust TUI — watch, steer, stop, resume > 2x cheaper, 2x faster than Browser Harness > GPT, Sonnet/Opus, Kimi, GLM, Qwen, DeepSeek I’ve used it for SF parking permits and Azure admin permissions. It asked me to login, then finished end-to-end. What are you going to use it for? Automate the boring stuff in the browser. 🔥 100% open source ↓

@GergelyOrosz And the low amount of MLE impacted means a shift in the eng skill distribution

@GergelyOrosz Plus the management flattening and EM forced to become IC

Shopify's Head of Engineering: "If you don't figure out how to harness agents in 2026, you'll be behind." This interview is the most practical breakdown of enterprise AI coding I've seen this year. Farhan Thawar explained the full Shopify AI playbook here. Watch the interview, then grab the exact template below 👇

darkzodchi @zodchiii

a month ago

x.com/i/article/2056…

26 62 650 813K 3K