The Five-Layer Cake
Five layers stand between electricity and intelligence — energy, chips, infrastructure, models, applications. Each one gates the layer above it.
← All insights
Framework · Q2 2026 Letter
Agent = LLM + Harness
A model is not an agent. What turns capability into completed work is the scaffolding around it — and that is where the value accrues.
Building on the five-layer cake AI illustration detailed last quarter, the basic operating cycle of an AI agent is shown in the illustration below. I believe billions of digital and physical agents will be operating in the near future, requiring massive amounts of compute, networking, storage, and cybersecurity to operate around the clock.
This diagram showcases a harness, which is the operating system around an LLM. The harness — composed of prompts, memory, tools, security, orchestration, and feedback loops — turns raw model intelligence into reliable actions by observing context, constraining behaviors, invoking tools, remembering state, and executing workflows.
The prompt defines the user’s intent or objective; then the agent gathers context from memory, data, tools, files, and applications; observes the current state of the task and any changes or constraints; reasons through what should happen next, including trade-offs, security & governance rules, and required tool calls; and then acts by executing a step, producing an output, invoking software, updating memory, and orchestrating the next workflow action. The result then feeds back into the loop until the objective is complete. All this can now be done without human intervention via loops and automations. When you leave your desk for the night, your AI agents will work simultaneously while you sleep to produce work that would have taken days, weeks, or months to complete. This continuous feedback loop and around the clock usage is driving massive new compute requirements even higher, accelerating in a vertical, exponential trajectory of demand that far outstrips available supply.
Harness component 1 of 5Prompt
The prompt is the instruction set that’s either typed in or verbally delivered by the user. It defines what the agent is trying to do, the constraints it must respect, and the policies it cannot violate. In the agentic era this is no longer a single string but a managed, versioned, evaluated artifact, where system prompts, tool descriptions, and policy scaffolding are rolled out like production code and scaled across organizations. This is the current interface with AI models, but it will evolve as more applications will be integrated at this layer.
More frameworks
Sovereign AI
Compute became strategic infrastructure. Nations began buying it the way they once bought refineries and fabs.
Disclosures
This page excerpts the framework discussion from Katam Hill LLC’s Q2 2026 Letter to clients, dated July 6, 2026. It omits portfolio performance and reflects views as of that date, not as of any later date. Certain economic and market information contained herein has been obtained from published sources prepared by other parties. While such sources are believed to be reliable for the purposes used herein, Katam Hill LLC does not assume any responsibility for the accuracy or completeness of such information. Unless otherwise indicated, market analysis and conclusions are based upon opinions or assumptions that Katam Hill LLC considers to be reasonable.
Securities or companies highlighted or discussed on this page have been selected to illustrate Katam Hill LLC’s investment approach and/or market outlook and are not intended to represent the adviser’s performance or be an indicator for how client accounts have performed or may perform in the future. Each has been selected solely for this purpose and not on the basis of performance or any performance-related criteria. Clients’ accounts are actively managed, and securities discussed may or may not be held in such portfolios at any given time. The securities referenced do not represent all of the securities purchased, sold, or recommended to clients. A complete list of all recommendations made by Katam Hill LLC within the past twelve months is available upon request.
Nothing on this page shall constitute advice or a recommendation or offer to sell or a solicitation to deal in any security or financial product. It is provided for information purposes only. Readers should not rely on this material in making any future investment decision. Past performance is not indicative nor a guarantee of future results. Actual returns for a particular investor will vary.