These agents are focused, private and they all run on my desktop. This wasn’t possible 1 - 2 years ago. The models weren’t good enough. The hardware wasn’t good enough. This dramatically changed in the beginning of 2026 and the latest improvements for local models keep coming.

What I’m realizing as I work with these models more is that they simply do not need to be trillion parameter frontier models in order to accomplish the majority of my tasks. In fact, a mediocre model with the right tooling is often better than a frontier model.

How tools are used

When prompted “What are my tasks for today?”, Jarvis —using gemma4:26b— scans the tools accessible to him. Each tool is a small python function that can be used to accomplish a goal. The model knows it needs to find “today” from the list of date blocks in my main markdown file, so it uses fetch_dates tool which returns an array of all my content in a sorted array. Once it finds the date block associated with today, it then calls fetch_tasks_from_dates(today_block) and returns an array of all tasks.

Why do we need to have explicit code to perform this action? Can’t something like Claude or ChatGPT handle this with only prompts? Probably. Frontier models are exceedingly good at complex tasks. When I attempted to do this locally with only prompts, I kept getting inconsistent results. The model would trip up on showing only incomplete or completed tasks because the prompt was requesting tasks across multiple dimensions (state, date). Once I introduced the tools to accomplish the task with code, the local model has never failed since.

Where it gets very interesting (and dangerous!) is allowing Jarvis to write his own tools. Since python is dynamically interpreted, I could even allow it to execute on the fly tools through eval(). In laymen terms, Jarvis could create the python tool and run it without saving it at all; the code is created and executed on the fly. I have not yet done this mostly because I don’t think the local models are good enough, but I expect that they will hit that capability within the next year.

Multiple models and a classifier

Jarvis is recursive. He also classifies every message I send using a tiny gemma model, then routes accordingly. If I ask a medical question, he will call himself but bypass the default model and select medgemma:27b, an open-source medical model that can answer medical questions and analyze medical imagery. If I ask him a deep reasoning question, he will do the same but call qwen3.6. This allows me to have small, focused models that are excellent at specific tasks.

Why do I need frontier again?

I still use Claude Code for work, but I’m finding myself increasingly uninterested in frontier models for anything outside my professional life. I rarely send personal information out of my local network, so I’ve never been able to use frontier models for things like medical questions or analysis of personal information. Now I can just do all of this with my local setup. Here’s a non-exhaustive list of what I’m currently doing with local models:

• Jarvis will use mental models like inversion or second-order thinking to debate me on topics and force me to think through issues I’m dealing with

• If I ask a question that requires deeper research, Jarvis will call Jones and wait for them to finish, then present me with the synthesized report

• If I add “QUESTION: ….” in my main markdown file, Jarvis will periodically scan the file for any questions that are unanswered, then attempt to find the answer and add it to the file below as “ANSWER: …”

• Jarvis is both a CLI tool and a full TUI. I can call him in the command line and get a response, or load a full fledged interface

• When I open my mail, I process it with a ScanSnap scanner and discard the physical document. It’s then uploaded to Paperless-ngx and processed using glm-ocr for OCR text, categorized and a title set by gemma4:26b and then associated with the correct corespondent.

Frontier models will always have their place, especially in enterprise settings, where speed matters and in specific domains where generalized knowledge of everything is important, but for personal knowledge management or routine tasks around the home, I cannot imagine using anything other than the models I have under my control.

Local models will keep getting better and cheaper. Apple shipped a (albeit expensive) laptop this past few months that can run very large models as well as my desktop. The hardware I had to buy now to do this will be obsolete in a year or two. I believe companies like Anthropic and OpenAI are moving aggressively into the enterprise space because they need to lock in contracts with big companies and provide guardrails appealing to businesses; consumers will eventually just rely on local models provided by Google or Apple on their personal hardware without needing to pay a hefty subscription fee to the big AI companies.

If our species was farther along in our evolution, the device sent to the alien planet might do far more than transform a planet, it might alter their solar system or even their galaxy, depending on the depth of knowledge contained on the device. It could teach them how to build spaceships that can travel vast distances, construct Dyson spheres, or show them how to build universal printers that can transform floating hydrogen atoms into anything they required.

A single tiny device encoded with only knowledge could transform an entire galaxy. That’s the power of knowledge.

With the recent release of powerful LLMs to the general public, I’ve found myself thinking a lot about how knowledge is acquired and used in our world today. LLMs have enabled millions of people to search for information without vetting the source or needing to build a larger model of the information from disparate sources, as they once had to do with books and traditional web searching. The analogy that comes to mind is that we were once students in a library, researching topics of interest; now we are simply asking the professor for the answer.

Why is this a problem? Shouldn’t getting to the answer by the shortest path possible be more useful and productive? For information, yes, it’s quite useful. Quickly determining if a tornado is coming to your town or if your favorite sports team is winning is a great use of this technology. For knowledge —useful information that is productive— it’s more nuanced because of its role in the creation of deep insights.

Deep insight is the creation of new, profound knowledge by combining existing knowledge in creative ways. It is what is behind every breakthrough in human history. The discovery of relativity by Einstein, Ford’s invention of the assembly line and Satoshi Nakamoto’s Bitcoin whitepaper are examples of new, profound knowledge that alters history. These insights are rare; most people create few in their lifetimes and some never do. That is because to create a deep insight not only requires a certain amount of knowledge stored in memory, but also the creativity necessary to combine/deconstruct it into new knowledge. Having creativity without knowledge is not enough; the most brilliant mathematician born into this world will not transform the field of mathematics if they are illiterate and thousands of miles from civilization.

In contrast, shallow insights create no new knowledge (unbeknownst to the creator) or knowledge that is inconsequential. People frequently produce shallow insights —young adults often do this, believing they have discovered something novel, only to realize someone else has already stumbled upon their profound realization. More importantly, LLMs only produce shallow insights due to how they work. LLMs are a very specific kind of artificial intelligence that predicts the next token from a sequence of previous tokens. LLMs do not have creativity like humans, and yet this does not stop many individuals from overselling their capabilities.

LLMs are indeed powerful and useful when working with existing knowledge. Their ability to summarize, aggregate and query is impressive. Consuming thousands of websites on the topic of migratory birds and producing a small paragraph summary is a great use case for LLMs. In addition, they are quite powerful at iterating upon or creating variations of existing knowledge. However, they’re ability to create new and profound knowledge (deep insight) is not possible. LLMs are not on the path to AGI, regardless of biased CEOs proclamations.

Now that we understand LLMs only produce shallow insights, why is it a problem for us to rely on them when they are in fact quite good at querying and varying existing knowledge? To understand this, we must remind ourselves again what is required for deep insight:

• existing knowledge in memory (preferably varied)

• creativity

A creative person without what I call minimum viable knowledge (MVK) stored in memory cannot produce deep insight. There simply isn’t enough material for the creativity to build connections upon. This is why young children and destitute, illiterate geniuses don’t produce world-changing inventions and breakthroughs; they lack knowledge. MVK is the boring blocks of knowledge that create more interesting clusters in our brains. Multiplication, basic handling of paint brushes, foundational equations in physics are all examples of the kinds of basic knowledge required to create insights in different fields.

To lean on LLMs for our minimum viable knowledge means we can externally store vast amounts of knowledge and access it faster but lose the ability to create deep insights. LLMs remove the friction of that painful, tedious learning of fundamental knowledge but at great cost. Without deep insights, we stagnate and our knowledge does not expand.

As a society, we must be aware of the dangers of relying too heavily on tools and agents that reduce our ability to acquire minimum viable knowledge. This is especially true for children and adults at the early stages of their careers or studies. Our over-reliance on nascent AI technology will reduce our ability to produce deep insights and until there is a breakthrough in artificial intelligence, we are the only known entities that can produce deep insights.

Here’s a parting question:

Imagine that LLMs were discovered in 1900 and trained on all human knowledge up to that point in time. Would the LLMs have been able to conceive of Einstein’s relativity before him?

Another attribute of the physical world that is key to understanding scarcity is the requirements for replication: resources and knowledge. In order for me to replicate an object in real-life, I must have the knowledge necessary to reconstruct it and the raw resources to build it. If I see a beautiful, hand crafted Japanese sword that I want, I not only need to understand how to make it, I would also need the raw materials. Only one does not get me the sword. If I am gifted in courtship (knowledge) but alone on an island (lacking resources), I will not fall in love. The physical universe is very good at constraining resources.

With this understanding, we can see how powerful figures arose throughout history simply by hoarding raw materials. If you control resources, you control production of valuable, necessary items for survival. Even if your enemies have the requisite knowledge to build machines of war, if you control the supply of necessary metals, they are not a threat.

In the past, replication was constrained by access to resources and knowledge because we operated almost entirely in physical space and our technology was limited. The future will be different. We will operate more frequently in virtual spaces, regardless of our preferences. Knowledge workers will conduct meetings and sessions in virtual offices. Children will learn in virtual classrooms. People will meet, fall in love and have relationships without ever meeting physically.

VR is not constrained by physical laws like our real-world. It is constrained by energy and computation (technically storage as well, but storage is cheap and getting cheaper so quickly that its cost will be negligible). It takes energy to power the computations necessary to operate VR and it requires CPUs/GPUs to render it. The latter is technically constrained in the physical world, but every year chips get faster, cheaper and more commoditized. Access to technology is so pervasive, even some of the poorest people on this planet have smart phones with advanced chip-sets. This leaves us with energy, which is also physically constrained. However, energy production is becoming increasingly more efficient and as technology improves —especially solar and battery technology— we will soon have access to virtually unlimited energy by capturing the majority of the solar energy hitting the Earth or by fusion.

Energy and computation only constrain VR externally. Internally, VR has no constraints other than knowledge. The concept of limited raw materials is nonsensical in most virtual environments. If I draw in Photoshop, I do not worry about running out of the color red. If I’m building a 3D model, there’s no shortage of polygons. Scarcity is not a concern.

In addition to essentially unlimited raw materials, objects in virtual environments can be perfectly replicated with only knowledge. This is already a large problem in present digital systems. The music industry was turned upside down in the early 2000s because replicating owned music moved from requiring knowledge and resources (the music files and the machinery to produce CDs) to just knowledge (knowing where to download the file). This same problem has plagued every form of media, including film and digital art.

Digital systems have unlimited resources and perfect replication with only knowledge. This is a problem.

If the future is virtual spaces and the metaverse, and these environments have unlimited raw materials, how can economies and valuable items arise without scarcity? If generating income moves into VR, what would compel someone to train and build amazing new creations if their hard labor was instantly reproducible? Economies will not emerge without scarcity.

“But wait, we already have scarcity in digital systems! Look how video games and social networks create economies and enforce scarcity.” This is true, closed systems can create artificial scarcity by restricting replication and controlling the “resources” of their ecosystem. However, closed systems are dangerous for two reasons.

First, they control how the resources work and can change the rules at any moment. Imagine being a laborer inside a closed system controlled by a corporation. You’ve built up a business inside their virtual world and are making money mining a resource that is hard to find. One day you wake up and discover a new software patch the corporation released has changed the quantity of that resource considerably and now the resource is abundant. There’s nothing you can do; the rules of the environment have simply changed.

Second, closed systems are ephemeral. Corporations shutter. Business owners trade hands and change direction. There is absolutely no guarantee any business you currently work with will be around in 10, 5 or even one year. So not only are the parameters of a closed system unreliable, there’s no guarantee that the system will persist.

Closed systems —Fortnite, Facebook, TikTok, etc— were up until recently the only real way to restrict resources and replication inside digital environments. Fortunately, that is no longer the case. With the invention of blockchain technology, we can now have an open system that allows for the persistent storage of digital assets and verified scarcity.

This is a revolution that’s for the most part ignored by the majority of people. For the first time in the history of modern computers, a means to verify scarcity exists in a system not controlled by a private corporation. You can own a unique digital asset that no one else has. Everyone can verify only you own it. Even if copies exist elsewhere, everyone will know your version is legitimate. Most importantly, this asset will not disappear when the business/person/entity that made it disappears. As long as the blockchain persists, the object persists, regardless of the businesses that come and go.

From my view, this explains the fever induced explosion in NFT prices from 2021 on. From a cursory view, it may appear to be strange and bubbly (and it most certainly is bubbly), but the people invested in this space see the potential and believe they are buying some of the first assets ever to be part of this pivotal moment in history.

There is no future in virtual systems without scarcity and there are only two known ways to implement it that we know of: proprietary, closed systems and blockchain technology. The former is closed, unreliable and ephemeral. The latter is open, reliable and has the potential to last indefinitely. Which will we choose?