has maybe been a very powerful phrase on the subject of Massive Language Fashions (LLMs), with the discharge of ChatGPT. ChatGPT was made so profitable, largely due to the scaled pre-training OpenAI did, making it a robust language mannequin.
Following that, Frontier LLM labs began scaling the post-training, with supervised fine-tuning and RLHF, the place fashions acquired more and more higher at instruction following and performing advanced duties.
And simply after we thought LLMs have been about to plateau, we began doing inference-time scaling with the discharge of reasoning fashions, the place spending considering tokens gave enormous enhancements to the standard of outputs.
I now argue we should always proceed this scaling with a brand new scaling paradigm: usage-based scaling, the place you scale how a lot you’re utilizing LLMs:
- Run extra coding brokers in parallel
- At all times begin a deep analysis on a subject of curiosity
- Run data fetching workflows
Should you’re not firing off an agent earlier than going to lunch, or going to sleep, you’re losing time
On this article, I’ll talk about why scaling LLM utilization can result in elevated productiveness, particularly when working as a programmer. Moreover, I’ll talk about particular methods you need to use to scale your LLM utilization, each personally, and for corporations you’re working for. I’ll maintain this text high-level, aiming to encourage how one can maximally make the most of AI to your benefit.
Why it is best to scale LLM utilization
Now we have already seen scaling be extremely highly effective beforehand with:
- pre-training
- post-training
- inference time scaling
The explanation for that is that it seems the extra computing energy you spend on one thing, the higher output high quality you’ll obtain. This, after all, assumes you’re capable of spend the pc successfully. For instance, for pre-training, having the ability to scale computing depends on
- Massive sufficient fashions (sufficient weights to coach)
- Sufficient knowledge to coach on
Should you scale compute with out these two parts, you gained’t see enhancements. Nonetheless, for those who do scale all three, you get wonderful outcomes, just like the frontier LLMs we’re seeing now, for instance, with the discharge of Gemini 3.
I thus argue it is best to look to scale your individual LLM utilization as a lot as attainable. This might, for instance, be firing off a number of brokers to code in parallel, or beginning Gemini deep analysis on a subject you’re taken with.
In fact, the utilization should nonetheless be of worth. There’s no level in beginning a coding agent on some obscure process you haven’t any want for. Somewhat, it is best to begin a coding agent on:
- A linear subject you by no means felt you had time to take a seat down and do your self
- A fast function was requested within the final gross sales name
- Some UI enhancements, you realize, right now’s coding brokers deal with simply
In a world with abundance of sources, we should always look to maximise our use of them
My major level right here is that the edge to carry out duties has decreased considerably because the launch of LLMs. Beforehand, if you acquired a bug report, you needed to sit down for two hours in deep focus, desirous about learn how to clear up that bug.
Nonetheless, right now, that’s not the case. As a substitute, you possibly can go into Cursor, put within the bug report, and ask Claude Sonnet 4.5 to aim to repair it. You may then come again 10 minutes later, take a look at if the issue is fastened, and create the pull request.
What number of tokens are you able to spend whereas nonetheless doing one thing helpful with the tokens
scale LLM utilization
I talked about why it is best to scale LLM utilization by working extra coding brokers, deep analysis brokers, and another AI brokers. Nonetheless, it may be arduous to think about precisely what LLMs it is best to hearth off. Thus, on this part, I’ll talk about particular brokers you possibly can hearth off to scale your LLM utilization.
Parallel coding brokers
Parallel coding brokers are one of many easiest methods to scale LLM utilization for any programmer. As a substitute of solely engaged on one downside at a time, you begin two or extra brokers on the identical time, both utilizing Cursor brokers, Claude code, or another agentic coding instrument. That is usually made very straightforward to do by using Git worktrees.
For instance, I usually have one major process or venture that I’m engaged on, the place I’m sitting in Cursor and programming. Nonetheless, generally I get a bug report coming in, and I mechanically route it to Claude Code to make it seek for why the issue is going on and repair it if attainable. Typically, this works out of the field; generally, I’ve to assist it a bit.
Nonetheless, the price of beginning this bug fixing agent is tremendous low (I can actually simply copy the Linear subject into Cursor, which may learn the difficulty utilizing Linear MCP). Equally, I even have a script mechanically researching related prospects, which I’ve working within the background.
Deep analysis
Deep analysis is a performance you need to use in any of the frontier mannequin suppliers like Google Gemini, OpenAI ChatGPT, and Anthropic’s Claude. I desire Gemini 3 deep analysis, although there are numerous different stable deep analysis instruments on the market.
Every time I’m taken with studying extra a few subject, discovering data, or something comparable, I hearth off a deep analysis agent with Gemini.
For instance, I used to be taken with discovering some prospects given a selected ICP. I then rapidly pasted the ICP data into Gemini, gave it some contextual data, and had it begin researching, in order that it may run whereas I used to be engaged on my major programming venture.
After 20 minutes, I had a short report from Gemini, which turned out to include a great deal of helpful data.
Creating workflows with n8n
One other solution to scale LLM utilization is to create workflows with n8n or any comparable workflow-building instrument. With n8n, you possibly can construct particular workflows that, for instance, learn Slack messages and carry out some motion primarily based on these Slack messages.
You might, as an example, have a workflow that reads a bug report group on Slack and mechanically begins a Claude code agent for a given bug report. Or you would create one other workflow that aggregates data from plenty of completely different sources and offers it to you in an simply readable format. There are primarily limitless alternatives with workflow-building instruments.
Extra
There are a lot of different methods you need to use to scale your LLM utilization. I’ve solely listed the primary few objects that got here to thoughts for me once I’m working with LLMs. I like to recommend all the time retaining in thoughts what you possibly can automate utilizing AI, and how one can leverage it to develop into simpler. scale LLM utilization will fluctuate broadly from completely different corporations, job titles, and lots of different components.
Conclusion
On this article, I’ve mentioned learn how to scale your LLM utilization to develop into a simpler engineer. I argue that we’ve seen scaling work extremely properly prior to now, and it’s extremely doubtless we will see more and more highly effective outcomes by scaling our personal utilization of LLMs. This may very well be firing off extra coding brokers in parallel, working deep analysis brokers whereas consuming lunch. On the whole, I consider that by rising our LLM utilization, we will develop into more and more productive.
👉 My free eBook and Webinar:
📚 Get my free Imaginative and prescient Language Fashions book
💻 My webinar on Imaginative and prescient Language Fashions
👉 Discover me on socials:
🧑💻 Get in contact
✍️ Medium
