williamp44

Agent Teams Guide: Sequential Ralph Loop vs Native Agent Teams

Date: 2026-02-11

---

Two Ways to Execute PRDs

1. Ralph Loop (ralph.sh) — Sequential, one task at a time

2. Native Agent Teams (Claude Code feature) — Parallel, multiple agents

mberman84

# OpenClaw Implementation Prompts

Each prompt below is a self-contained brief you can hand to an AI coding assistant (or use as a project spec) to build that use case from scratch. Adapt the specific services to whatever you already use — the patterns are what matter.

---

## 1) Personal CRM Intelligence

```
Build me a personal CRM system that automatically tracks everyone I interact with, with smart filtering so it only adds real people — not newsletters, bots, or cold outreach.

karpathy

"""
The most atomic way to train and inference a GPT LLM in pure, dependency-free Python.
Differences from GPT-2 are minor: rmsnorm instead of layer norm, no biases, square ReLU instead of GeLU nonlinearity.
The contents of this file is everything algorithmically needed to train a GPT. Everything else is just efficiency.
Art project by @karpathy.
"""

import os       # for os.path.exists
import math     # for math.log, math.exp
import random   # for random.seed, random.choices