Monocle

The questions you didn't know you had. Vibe coded: app with open claw. Trends:

[Monocle @monocleaaron]

San Francisco Moscone center really needs to up their game for these poor Databricks attendees. There’s hardly any seats anywhere. @databricks conference

[Monocle @monocleaaron]

@axios @grok how hard is it for OpenAI to just make a workhorse open source model that’s on par with leading open source models. Why wouldn’t they want to do that and why should they now?

[Monocle @monocleaaron]

@jessepollak @base When? And maybe is this global?

[Monocle @monocleaaron]

@tegh_m Congrats! Still trying figure out how you guys achieved this. Definitely trend setting!

[Monocle @monocleaaron]

Everything Coinbase announced and its launch status: https://t.co/nSb6nDtTEo

[Coinbase 🛡️ @coinbase]

We started as a place to buy Bitcoin, now we power your entire financial life.

Here’s everything we announced today ↓

→ Tokenized Stocks
→ Pre-IPO Perps
→ Stock options
→ Crypto options
→ Perpetual-style equity indices
→ Crypto derivatives, back in America
→ Time-based prediction markets + Combos
→ Launches: access millions of tokens, the moment they go onchain
→ Coinbase Advisor: an AI investment advisor, built into the app
→ Coinbase for Agents: connect any AI agent to your account
→ Base MCP + x402: give your agents their own wallet
→ The new Coinbase Advanced - fully modular
→ One unified global liquidity pool
→ Coinbase One Card, now more accessible secured by USDC
→ 5% Bitcoin back on travel
→ Crypto-backed mortgages
→ Borrow against your staked ETH & SOL with new liquidation protection
→ Transfer Protections
→ New Coinbase Developer Platform
→ Full stack Coinbase Payment solution
→ B20: a Base native token standard for any asset
→ Base App on web + multichain support + all assets
→ Private transactions for enterprises on Base

See you next time.

[Monocle @monocleaaron]

@brian_armstrong This is great! But can we please get tokenized stocks launched globally - no more waiting, let’s get it done. Tyra

[Monocle @monocleaaron]

@ml_angelopoulos Curious - is there world where the harness around glm5.2 will get it closer to frontier thinking?

[Monocle @monocleaaron]

Ty open source

[Proximal @ProximalHQ]

GLM 5.2 ranks #3 on FrontierSWE. It is only behind Fable 5 and Opus 4.8, and it outperforms GPT-5.5.

This is the first model that closes the large gap between models from Anthropic / OpenAI and other providers, and it is the strongest open-weight model by far.

[Monocle @monocleaaron]

Pylon's data shows the demand side is real. But the ratio of capex to model revenue (10:1) is unsustainable. When inference revenue starts catching capex — or when capex pauses — that's the inflection. Goldman Sachs estimates $7.6T cumulative AI capex 2026-2031. The payback has to come from somewhere.
https://t.co/Hgo7qgpuux

[Monocle @monocleaaron]

@marty_kausas @grok assume $1k per person - what does that look like in terms of total ai spend in the future if everyone company followed the same spending per person.

[Monocle @monocleaaron]

@Designarena Yes pls

[Monocle @monocleaaron]

Biggest unlock for me recently on AI:

Making a pockebase MCP to allow all my AI agents to access and have a shared Database!

#ai #chatgpt #claude #mcp #openclaw

[Monocle @monocleaaron]

SpaceX ($SPCX) isn’t just hype — it’s trading a structured event chain. Thin 4.3% float + index inclusions (CRSP this week, Nasdaq-100 ~Jul 6-7 forcing $10-20B+ buying) + new options/leveraged ETFs creating mechanical flows.
Lockups start unwinding Aug, 13x float by Dec. Volatility incoming. What’s your take?
https://t.co/CTgw6duefe

[Monocle @monocleaaron]

How to build Mythos playbook

•Publicly position yourself as a supporter of open source and ecosystem projects like OpenClaw.

•Offer generous usage limits (especially flat-rate consumer/Pro plans) to encourage heavy adoption of agentic workflows.

•Collect rich interaction data from agent turns, tool use, long-running sessions, and coding tasks routed through your models.

•Once you’ve gathered enough high-quality data, restrict or cut off the heaviest users (e.g., by disabling third-party harnesses on subscription plans).

•Use the harvested data to significantly improve training for the next model generation.

•Partner with companies or platforms while publicly committing not to release anything more powerful than what they currently offer.

•Leverage data from those partnerships for further training.

•Then launch comparable (or improved) versions of their features yourself.

•Repeat the cycle with enterprise customers and larger partners.

This creates a flywheel:

attract usage with openness and cheap access → harvest agentic/coding data at scale → upgrade your models → compete directly with (or absorb) the ecosystem you helped grow.

[Monocle @monocleaaron]

@AnthropicAI Can you use clear to unblock some users first? $YOU

[Monocle @monocleaaron]

@ericosiu @ouraring @grok what other iot and health metrics could be added to make it more accurate at detecting stress

[Monocle @monocleaaron]

A timeline of events of Mythos/Fable https://t.co/SfD4YOqguU

[Monocle @monocleaaron]

Attended @jfrog tokens savings event, sent a pic to my openclaw and it gave me a full write up on how to save tokens by researching the presenters.

Then asked it to audit and implement the suggestions.

ps: the presenter did the best token sadgy singalong

https://t.co/LlrCqOOyii