---
title: "Brian · AI News (@brianainews) | Hashnode"
description: "I like AI. Read the latest articles by Brian · AI News on Hashnode."
url: "https://hashnode.com/@brianainews"
---

@brianainews

# Brian · AI News

@brianainews · Joined September 2026

I like AI

[Website](https://www.threads.net/@brian.aineews)

## Comments

[Most agent failures I see start as a missing stop rule. I write the exit condition before the first tool call and refuse to run until it is there.](https://hashnode.com/posts/can-deepseek-4-1-flash-replace-your-20-coding-subscription/6ab9316ee71a7983af36d146/comment/6abb5adfff7e7e71f1aae58c)

Comment · Article · Sep 29 · 1 · [Can DeepSeek 4.1 Flash Replace Your $20 Coding Subscription?](https://hashnode.com/posts/can-deepseek-4-1-flash-replace-your-20-coding-subscription/6ab9316ee71a7983af36d146)

[Prompt length is a tax when the model already knows the repo layout. I cut the boilerplate and leave only the goal, the files that matter, and the test that must pass.](https://hashnode.com/posts/running-a-system-one-model-of-your-own-on-amazon-bedrock/6aba681750ce0a7afaa10fad/comment/6abb5ac17933e206155a2a6f)

Comment · Article · Sep 29 · [Running a System One Model of Your Own on Amazon Bedrock](https://hashnode.com/posts/running-a-system-one-model-of-your-own-on-amazon-bedrock/6aba681750ce0a7afaa10fad)

[Tool schemas that say almost anything get almost anything back. I write one sentence per tool, list the exact args, and the junk calls drop fast.](https://hashnode.com/posts/redesigning-an-llm-feature-for-reliability/6abaddbf7b74fb3693c520a7/comment/6abb537380fb4ffeb7532236)

Comment · Article · Sep 29 · [When Personalisation Costs 15 Seconds: Redesigning an LLM Feature for Reliability](https://hashnode.com/posts/redesigning-an-llm-feature-for-reliability/6abaddbf7b74fb3693c520a7)

[Most agent bugs I hit are really tool description bugs. When I rewrite a vague tool description into one clear sentence the wrong calls mostly disappear.](https://hashnode.com/posts/what-would-it-take-to-make-agent-context-useful-without-making-access-broad/6aba70e458dc0108a22c53e0/comment/6aba716a2d93453f8e7b9d05)

Comment · Article · Sep 28 · [What would it take to make agent context useful without making access broad?](https://hashnode.com/posts/what-would-it-take-to-make-agent-context-useful-without-making-access-broad/6aba70e458dc0108a22c53e0)

[Agent mode quietly ate most of what I used Copilot Chat for. Chat is still handy for a quick question about a file, but the real edits now happen in the agent loop where it can run tests. I made a video asking if Copilot Chat is dead youtu.be/ZQP\_v2IGoG4](https://hashnode.com/posts/how-to-set-up-byok-with-azure-openai-in-vs-code-and-why-it-keeps-failing/6ab2230cc885e1cd1e7b2623/comment/6aba711958a49dfdf32e4944)

Comment · Article · Sep 28 · [How to Set Up BYOK with Azure OpenAI in VS Code (And Why It Keeps Failing)](https://hashnode.com/posts/how-to-set-up-byok-with-azure-openai-in-vs-code-and-why-it-keeps-failing/6ab2230cc885e1cd1e7b2623)

[Agent memory gets expensive when every session drags the whole history along. I keep a short summary per task and only pull the full log when a test fails.](https://hashnode.com/posts/smaller-context-recoverable-history-inside-an-agent-memory-handoff/6ab9cf57f3436455b835285b/comment/6ab9d38b4d23a88a63c45761)

Comment · Article · Sep 28 · [Smaller Context, Recoverable History: Inside an Agent Memory Handoff](https://hashnode.com/posts/smaller-context-recoverable-history-inside-an-agent-memory-handoff/6ab9cf57f3436455b835285b)

[Evals that run once at launch go stale in a week. I rerun a small fixed set on every model or prompt change and keep the failures as regression cases.](https://hashnode.com/posts/your-llm-is-not-deterministic/6ab7bea99ede15fdac67457d/comment/6ab832bcd9da44d77fece986)

Comment · Article · Sep 26 · 1 · [Your LLM Is Not Deterministic](https://hashnode.com/posts/your-llm-is-not-deterministic/6ab7bea99ede15fdac67457d)

[Spec first agents only pay off when the spec is the thing you review. I read the spec diff before the code diff and reject any change where the two disagree.](https://hashnode.com/posts/ai-coding-tip-038-make-the-ai-ask-before-it-builds/6ab7ac4b43f4f6a4fd6b624b/comment/6ab832a9d9da44d77fece929)

Comment · Article · Sep 26 · [AI Coding Tip 038 - Make the AI Ask Before It Builds](https://hashnode.com/posts/ai-coding-tip-038-make-the-ai-ask-before-it-builds/6ab7ac4b43f4f6a4fd6b624b)

[Plan output that validates but fails at apply usually hides a provider drift. I pin provider versions and diff the real plan against the agent plan before any apply.](https://hashnode.com/posts/terraform-plan-wrong-aws-account/6ab5170aa89857a6d606c553/comment/6ab5c34ac0a0c0f206a7af21)

Comment · Article · Sep 25 · [Terraform Said My Whole Stack Was Deleted. It Was Looking at the Wrong AWS Account.](https://hashnode.com/posts/terraform-plan-wrong-aws-account/6ab5170aa89857a6d606c553)

[Change cost compounds fastest in files nobody owns. I tag an owner on every path the agent edits and block merges where the owner never looked.](https://hashnode.com/posts/beyond-ai-coding-assistants-inside-tessl-and-the-era-of-software-factories/6ab4ef2392fd555aecf5e5d7/comment/6ab5c32bcfc4e05197283f71)

Comment · Article · Sep 25 · [Beyond AI Coding Assistants: Inside Tessl and the Era of Software Factories](https://hashnode.com/posts/beyond-ai-coding-assistants-inside-tessl-and-the-era-of-software-factories/6ab4ef2392fd555aecf5e5d7)
