Sharing a "RAM" with my agent: a stack-based memory keeps revisions on goal
alfcnz · x · 2026-11-18
alfcnz shares his personal workflow of sharing a "RAM" between himself and his AI agent. They use a stack structure: the ultimate goal sits at the bottom of the stack, so revision detours (formula tweaks, figure redraws) still converge on the original target. The screenshot shows the live stack: PyTorch chapter revision items, pending decisions like whether to use torch.setgradenabled(False), and parked items. He notes the more he builds his own zero-friction tooling around this, the better it works.
Related event: Dev Shares Stack-Based Shared RAM Workflow for Human-Agent Collaboration(2 posts)→
More from coding & agent
- aivoicebus 0.16.0 adds Grok Bot to its AI agent fleet via webhook — baditaflorin · 2026-11-18
- deadweight: a Rust CLI that measures wasted tokens in your Claude Code setup — JeremyCMorgan · 2026-11-18
- Shadow MCP services need governance: 'no policy means you're asking for failures' — colinmcnamara · 2026-11-18
- LangChain's GTM Agent: deals with agent-built sales docs advance 47% vs 16% — LangChain · 2026-11-18
- Trader wires Muse agents to execute CME futures trades, plans broker API for instant orders — CyborgTraderOFTAGI · 2026-11-18
- Qwen 3.8 27B vs Flash-Next vs DeepSeek V4.1 Flash: one-shot landing page build-off — jesdga95 · 2026-11-18