How I Build — Adam Montgomery
This is a guide to my tools and processes for building software. I'm writing this both to share what I've found with other humans, and as guidance for LLMs that interact with me & share my tools. I've found these tools helpful building "pre-LLM" software in small and large startups, and for LLM-enabled applications like I'm working on now. My tool choices are largely informed by how I structure software. I don't spend much time justifying my tools here. That's both for brevity, and because I think tools either resonate with a person or they don't . It's also not exhaustive—these are what I keep in the tool chest, not the attic.1 My primary development environment. Working roughly from the bottom up: I use a MacBook Air 15". I used to always go with a beefed up MacBook Pro, but switched and haven't missed it. If I do anything compute intensive, I usually just use another computer. I use Karabiner-Elements to map caps_lock -> left-ctrl, left-ctrl -> esc, and right-cmd + hjkl -> arrow key
This is a guide to my tools and processes for building software. I'm writing this both to share what I've found with other humans, and as guidance for LLMs that interact with me & share my tools. I've found these tools helpful building "pre-LLM" software in small and large startups, and for LLM-enabled applications like I'm working on now . My tool choices are largely informed by how I structure software . I don't spend much time justifying my tools here. That's both for brevity, and because I think tools either resonate with a person or they don't . It's also not exhaustive—these are what I k
saved by
related reading
- Building Effective AI Agents \ Anthropicanthropic.com
- crawshaw - 2025-01-06crawshaw.io
- Building Effective AI Agents \ Anthropicanthropic.com
- Malleable software in the age of LLMsgeoffreylitt.com
- Here’s how I use LLMs to help me write codesimonwillison.net
- Patterns for Building LLM-based Systems & Productseugeneyan.com
- Shipping at Inference-Speed | Peter Steinbergersteipete.me
- Andrej Karpathy on X: "LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating" / Xx.com
- llm-wikigist.github.com
- Your AI Product Needs Evals – Hamel's Blog - Hamel Husainhamel.dev
- Codifying a ChatGPT workflow into a malleable GUIgeoffreylitt.com
- GitHub - brexhq/prompt-engineering: Tips and tricks for working with Large Language Models like OpenAI's GPT-4.github.com