# intro machine (blogs) — RSS Amplifier Recent posts from the 1 feeds in the RSS Amplifier directory that cover intro machine. Page: Feed: --- ## [Guardrails Are More Than Prompt Safety](https://alexaitken.nz/blog/guardrails-are-more-than-prompt-safety/) _2026-06-19 · Alex Aitken · Alex Aitken_ The nightmare version is easy to imagine. It is 2am. OpsGenie is paging you. A user-uploaded PDF contained instructions the model should have ignored, and now the system has produced an answer that could expose something it should not expose. Maybe that sounds dramatic. But the underlying problem is not exotic. A receipt, invoice, comment, \[ \] ## [You Do Not Fix Hallucinations With One Prompt](https://alexaitken.nz/blog/you-do-not-fix-hallucinations-with-one-prompt/) _2026-06-12 · Alex Aitken · Alex Aitken_ In the previous post, I wrote about deciding which parts of an expense workflow should be automated first. Some decisions are better handled with ordinary code. Some need model judgement. Some should surface information for a reviewer without making the final call. Over time, the goal is to automate more of the workflow as the \[ \] ## [The Hard Part Was Deciding What to Automate First](https://alexaitken.nz/blog/the-hard-part-was-deciding-what-to-automate-first/) _2026-06-05 · Alex Aitken · Alex Aitken_ The ideal expense workflow is one you barely notice. An expense is captured automatically. The right policy is applied. The supporting evidence is checked. Obvious cases move through without anyone needing to think about them. A person only gets involved when something genuinely needs judgement. I hate spending time submitting expenses. Nobody wants to spend \[ \] ## [The Prompt Had Too Many Jobs](https://alexaitken.nz/blog/the-prompt-had-too-many-jobs/) _2026-05-29 · Alex Aitken · Alex Aitken_ Prompt engineering is often talked about like it s magic. Write the right words. Add the right examples. Tell the model to be careful. Tell it to think step by step. Tell it not to hallucinate. Tell it to output JSON. Tell it to be concise. Tell it to be accurate. Tell it to do everything. \[ \] ## [Intro to Machine Learning: Multiple Linear Regression](https://alexaitken.nz/blog/intro-to-machine-learning-multiple-linear-regression/) _2025-11-03 · Alex Aitken · Alex Aitken_ In the last post, we explored simple linear regression, where one feature predicts one target. That single line already gave us pretty good insights into relationships between variables. But real-world problems don t just depend on one factor. Let s take some examples: House prices are influenced by more than just size. Salary depends on more than \[ \] ## [Intro to Machine Learning: Simple Linear Regression](https://alexaitken.nz/blog/intro-to-machine-learning-simple-linear-regression/) _2025-10-27 · Alex Aitken · Alex Aitken_ When I first thought about machine learning, it seemed like a scary topic. Coming from product engineering, I often heard about models, black boxes, and mysterious choices around which model to use. It felt inaccessible and mathematical. Recently, I decided to go beyond that assumption. To my surprise, a lot of it builds on concepts \[ \] ## [Interviewing as an Engineering Leader](https://alexaitken.nz/blog/interviewing-as-an-engineering-leader/) _2025-01-30 · Alex Aitken · Alex Aitken_ I ve been through my fair share of engineering interviews over the past few years. I ve looked at large companies and smaller start-ups like companies. Everyone has a different process, but there s one thing in common: you need to have stories. Being a leader is being a storyteller. I ve added some tips below to help you \[ \] ## [Managing Low Performers](https://alexaitken.nz/blog/managing-low-performers/) _2025-01-23 · Alex Aitken · Alex Aitken_ It s hard when you have someone sandbagging your team. The underperformer is dragging everyone down with their performance but it s also hard to eliminate them. You know them, and you ve built up a rapport. You know about their family and their struggles over the past few years. Ideally, we shouldn t be here. Everyone should \[ \]