The skill of using AI well is shifting from prompts to loop design. Loop engineering, which rose in mid-2026, is about designing the loop an agent runs on its own, instead of instructing the agent step by step. Andrej Karpathy ran a coding agent for two days, conducted 700 experiments, and found 20 optimizations that sped up training. ASAP summarizes the shift from the primary remarks.

After prompts comes the harness, then the loop

The skill of working with AI is now in three stages: prompt, harness, and loop. First was prompt engineering, writing one good prompt. Next was harness engineering, filling the context window with the right information. Now it is loop engineering, designing the loop an agent runs on its own.

"The Karpathy Loop": 700 experiments in two days

The flagship case of loop engineering is Karpathy's experiment. Andrej Karpathy ran a coding agent nonstop for two days, conducted 700 experiments, and found 20 optimizations that cut training time. A human did not instruct each step; one well-designed loop repeated on its own.

AutoResearch: keep only the changes that win

AutoResearch is Karpathy's open-source tool that implements loop engineering directly. You describe research directions in a markdown file and point an agent at the repo, and the agent repeats experiments while keeping only the changes that beat the current best. The human just sets direction and walks away.

The loop does the work, not the model

The core of loop engineering is that results come from the whole loop, not one answer from the model. That is why tools like Codex are becoming agent loop managers rather than chatbots. Which loop you run matters more than what you ask.

The caveat: beware "loopmaxxing"

Running a loop endlessly, a habit the field calls loopmaxxing, is not the answer. A good loop needs stopping conditions, verification, and feedback to avoid the cost of spinning in place.

What it means: the new core skill is loop design

Loop engineering is the shift, in using AI well, from instruction to design. Someone who builds a loop that runs, verifies, and stops on its own gets more done than someone who asks one good question. It is the core skill after prompting.

Wrap-up

Loop engineering is the new skill of designing the loop an agent runs. It is the stage after prompts and the harness, and Karpathy's two-day, 700-experiment run shows its power. But you must avoid loopmaxxing and build in stopping conditions and verification.

Source: ASAP summary of Andrej Karpathy's "Loopy Era" and AutoResearch remarks (running a coding agent for two days, about 700 experiments, 20 training optimizations found) and 2026 reporting on loop engineering.