- Create a docker file for image:
# Set a common base
FROM alpine as base
# Set required CNB information
ENV CNB_USER_ID=1000| #/bin/bash | |
| set -e | |
| if [[ -n "$DEBUG" ]] | |
| then | |
| echo "!!! Debug output enabled" | |
| set -x | |
| fi | |
| set -u |
Two Execution Options
Subagent-Driven (this session) - I dispatch fresh subagent per task, review between tasks, fast iteration
Parallel Session (separate) - Open new session with executing-plans, batch execution with checkpoints
Which approach, Mark?
Mark wants more detail on the two execution options. Let me explain them clearly based on what the writing-plans skill says and my understanding of the superpowers skills.
| """ | |
| The most atomic way to train and run inference for a GPT in pure, dependency-free Python. | |
| This file is the complete algorithm. | |
| Everything else is just efficiency. | |
| @karpathy | |
| """ | |
| import os # os.path.exists | |
| import math # math.log, math.exp |