indubitably
Request a consultation

Engineering notes

Notes from the work.

AI systems, automation, and lessons learned building software from the Oregon coast.

  1. Rat Things: cloud agents in your AWS account

    A look at our current work on durable agent sessions, independent forks, and a native console — and the boundaries of the engineering preview.

  2. How to build your own Agentic Code Editor

    Build a tiny terminal coding agent in Python that chats with Claude and can read, list, and edit files. It's fast to set up and easy to extend. You'll go from zero to working in minutes.

  3. What are the Three Core Principles of DevOps?

    These three DevOps principles have been proven to dramatically improve the performance of technology organizations.

  4. Automate your Framer website with Make AI

    Every time we do something more than twice—automate. It's a rule of thumb of engineering in our humble opinion.

  5. Using AI To Help Read The 1012 Page Omnibus Bill

    I think we're all tired of these huge 'must pass' bills here in the United States and our elected officials are signing into law legal documents they themselves didn't even read.

  6. Use Claude3 To Understand Your Home Owners Association

    Do you feel like you need a law degree to understand your home owner's association contract when you buy a new house? You're not alone—but fear not there are tools out there to help you understand the contracts, rules, and regulations.

  7. The Frugal Architect: Making Cost A First-Class Concern

    Traditionally, cost has been an afterthought in system design and operations. But considering cost implications from the very start allows for more balanced trade-offs between features, time to market, and efficiency.

  8. Introducing CarnivoreGPT

    The team over at OpenAI ship code FAST! GPTs and the new Assistants API aim to help users of their platform easily create their own custom GPT.

  9. DALL·E 3: A Quantum Leap in Text-to-Image Generation

    The future of creative AI has arrived with OpenAI's DALL·E 3. This model understands nuance and detail like never before, translating your ideas into exceptionally accurate images.

  10. Mastering Prompt Engineering: Unlocking the Full Potential of AI Models

    In the era of powerful AI models like GPT-3 and Cohere, prompting is emerging as a new coding language that uses natural language instead of traditional symbols.

Have a project in mind?

Tell us what you want to build or improve. Greg replies to every inquiry personally.

Request a consultation