Adrian Scott | A.I. + Business Upscaling

Biodefense is dramatically more urgent than any climate issue. #biodefense #climatechange npr.org/sections/healt…

Banksy should make a statue of an Englishman being stabbed by a foreigner, or is that too realistic for his liking?

@ithavee Fascismo. Mediocre, y bien decel.

En Boquete planean que El que derribe árboles (por construcción dentro de tu terreno) debe sembrar la misma cantidad. Según veo, mucha gente está de acuerdo. Yo también!

@planbexpat @Ale_PPathways Not forced. Just bending over and simping.

You’re not wrong on the sovereignty angle. That tension is real. Panama is essentially being forced to legislate by Brussels to get off their blacklist. The irony is the reform is narrow enough that the core territorial tax system stays intact. But yes, the direction of travel is uncomfortable for anyone who values jurisdictional independence.

Panama is introducing a 15% corporate tax to all companies who can't show proof of substance and operations in the country. So much for "Panama never changes rules" 🤷‍♂️

@AvilaAngie_ Que otro alcalde ha bajado el presupuesto tanto?

Lo de la plantilla de los Asesores de 6k c/u de Mayer, lejos de enojarme me da tristeza, la gran mayoría le da igual porque en redes está subiendo “lo que hace” A este ritmo el siguiente alcalde pude ser hasta un loro con buen equipo de Redes.

@raulconte10 Raúl, el no bajó el presupuesto más que todos los alcaldes anteriores? Let the man cook

Mayer tiene 30 funcionarios ganando salario de viceministro: 6k al mes, 180k mensuales del bolsillo de todos. Viva la reactivación económica a costa del dinero público.

Panama regalando su soberanía y liderazgo

@KevronSutton @Ale_PPathways Way too high

@Ale_PPathways What are the odds of this passing?

@arandomguy808 @Ale_PPathways We've seen this movie before, mariposa inocente

@Ale_PPathways For companies making over $750 million a year. It's been discussed since 2024

@planbexpat @Ale_PPathways No, Panama is going to get screwed by the EU again, just like every time before. Regalando soberanía.

A few things being missed here. This is a proposal, not law. Still needs National Assembly approval. And the 15% applies to passive income generated inside Panama for companies without real substance. A company with zero Panamanian revenue has nothing to tax. Panama is doing this to get off the EU blacklist, not to chase expats. Territorial tax on foreign income stays untouched

@AndreBConte Veo la idea en alcantarillas por el riesgo de robo, pero en sillas es tóxico

Nunca estuve y sigo sin estar de acuerdo con utilizar plástico reciclado en cada cosa publica, como alcantarillas y sillas. Cuando lo dije hace 2 años me tildaron de hater, hoy dia se van dando cuenta del por qué. Una búsqueda en google nos hubiera ahorrado, mucho…

@Teknium Hot da*n, been dreaming about this for a while now 🔥🔥🔥🚀🚀🚀

Our first dive into Multi-Agent Coordination and Cooperation is here, with Hermes Agent Kanban Orchestrate tasks across multiple agent profiles and dependencies easily and visually. Achieve more. See the docs here: hermes-agent.nousresearch.com/docs/user-guid…

So, the J6ers, under Biden were rounded up like dogs, arrested, imprisoned, given unfair trials and sent to prison for years AND ANTIFA WAS NEVER TOUCHED. But now, under President Trump, Antifa continues wreaking havoc, assaulting people, destroying property, but there are no FBI raids, arrests, or accountability?!? WHY IS THIS ALLOWED!!???

This bill is KYC for browsing the internet. If you want to upload a driver's license or do a face scan to use the internet this bill is for you. If you don't, it's time to fight it. Get the word out.

@Dan_Jeffries1 Gotta primary Josh Hawley out of there.

@ReclaimTheNetHQ Gotta primary Josh Hawley. He really is an idiot.

Hawley's GUARD Act just passed committee 22-0. Every American would have to upload a government ID or submit to a face scan to use an AI chatbot. Even for asking for algebra help or fixing a billing issue. The framing is child safety but the result is a national ID system for talking to a computer. reclaimthenet.org/senate-panel-b…

Hey @David_J_Bier, how does the Cato Institute “Immigrants Are Better Than You” model factor in the lost wages of a working American mother of two stabbed to death by an illegal El Salvadorian?

@DavidSHolz Ser, that is not total cost

how did we make deepseek outperform opus 4.7? i've been thinking about why "open model bad at tool calling" is almost always a harness problem, not a model problem. context: spent the two days looking at billions of tokens in @CommandCodeAI (tb open source ai cli) using deepseek. I ended up writing a tool-input repair layer. the trigger was watching deepseek-flash fail on the simplest /review run, every shellCommand and readFile call bouncing back with a raw zod issues blob, the model unable to recover because the error wasn't in a form it could read. by the end deepseek v4 pro was beating opus 4.7 6/10 times on our internal evals. a few things i learned that feel general: 1/ the failure modes aren't random they're a small finite compositional set. across deepseek-flash, deepseek v4 pro, glm, qwen, the same four mistakes repeat almost exactly: - sending `null` for an optional field instead of omitting it - emitting `["a","b"]` as a json *string* instead of an actual array - wrapping a single arg in `{}` where the schema expected an array (an "empty placeholder") - passing a bare string where an array was expected (`"foo"` instead of `["foo"]`) four repairs, ~30-100 lines each, ordered carefully (json-array-parse must run before bare-string-wrap or `'["a","b"]'` becomes `['["a","b"]']`). that is the whole catalogue. when i hear "this open source model can't do tool calls" i now assume one of those four, and so far that's been right ~90% of the time. 2/ the funniest failure mode is also the most revealing. deepseek-flash, when asked to edit or write a file, sometimes emits the path as a *markdown auto-link*: filePath: "/Users/x/proj/[notes.md](http://notes. md)" our writeFile tool obediently trued creating files literally named `[notes.md](http://notes .md)` until we caught it. this is not a hallucination. it's the post-training chat distribution leaking through the tool boundary the model has been rewarded for auto-linking in conversational output, and is applying that prior in a context where it makes no sense. the fix is two regex lines that unwrap only the degenerate case where link text equals url-without-protocol real markdown like `[click](https://x .com)` passes through untouched. this is also conditioning of their own tools during RL which were different from all other tools we write and ofc can't predict. "tool confusion" is a more useful frame than "capability gap." the model knows how to format a path. it just hasn't been told clearly enough that this path is going to fopen, not into a chat bubble. so we encode that hint at the schema level `pathString()` instead of `z.string()` and the leak is plugged for every path field at once. 3/ the design choice that mattered was inverting preprocess-then-validate to validate-then-repair. my first attempt was the obvious one: a preprocessing pass that normalized inputs (strip nulls, parse stringified arrays, etc.) before zod ever saw them. it broke immediately, writeFile content that *happened* to be json-shaped got rewritten before it hit disk. silent corruption, easy to miss in a smoke test. then i made it less greedy - parse the input as-is. if it succeeds, ship it. valid inputs are never touched. - on failure, walk the validator's own issue list. for each issue path, try the four repairs in order until one applies. - parse again. on success, log `tool_input_repaired:${toolName}`. on failure, log `tool_input_invalid:${toolName}` and return a model-readable retry message. the structural insight here is: when you preprocess, you encode a prior about what's broken. when you let the validator complain first, the schema is the prior, and you only spend repair budget at the exact paths the schema actually disagreed at. the validator is doing the work of localizing the bug for you. it's the same shape as cheap-then-careful everywhere else try the fast path, fall back on evidence. (this also gives you per-tool telemetry for free. you can watch repair rates per (model, tool) and notice when a model regresses on a specific contract before users do.) 4/ shape invariants and relational invariants need different fixes. the four repairs above all handle shape problems wrong type, missing key, wrong container. but read_file had a *relational* invariant: "if you provide offset, you must also provide limit, and vice versa." deepseek kept calling `readFile({ absolutePath, limit: 30 })` and getting an `ERROR:` back. you can't fix this with input repair, because each field is independently valid the bug is in the relationship between them. so i taught the function the model's intent instead. `limit` alone → `offset = 0`. `offset` alone → `limit = 2000` (matches common read tool ops default). then surfaced the decision back to the model in the result: "Note: limit was not provided; defaulted to 2000 lines. To read more or fewer lines, retry with both offset and limit." no `Error:` prefix, so the tui doesn't paint it red. the model sees what we picked and can self-correct on the next turn if our guess was wrong. transparency over silent magic wins big. repair where you can. extend semantics where you can't. surface the choice either way. zoom out: a lot of what looks like model capability is actually contract design. a strict schema is a choice with a cost it filters out noise, but it also filters out recoverable noise from any model that hasn't memorized the exact json contract you happened to pick. the largest commercial models eat that cost invisibly and are linient on tool calling because they've seen enough of every contract during pretraining; open models pay it loudly and get dismissed for it. the harness is where you mediate between distributions. four small repairs (i'm sure more to follow as we have three more merging today), two regex lines for auto-links, one relational default, one prefix change. the model didn't change. the contract got more forgiving in exactly the places it needed to be. deepseek v4 pro now beats opus 4.7 6/10 times on our internal evals. imo "skill issue" applies to the harness more often than the model.

@teddypowday yes, but he blocked all of them

Is there really nobody in SF who can pull garry tan aside for an honest chat about what's happening to him

@Enterate_507 @AvilaAngie_

@adrianscottcom @AvilaAngie_ La inundación es en general, diversas áreas, no únicamente vía Argentina. Dejen de seguir la cantaleta de acuaman, que salió del cargo hace 7 años, y exijan respuestas a las autoridades actuales.

Llueve en un país tropical, acto seguido la ciudad se inunda por partes y es un caos vehicular… Bienvenidos a Panamá 🙃😢