Home

Recent posts

• What I have been reading this week (April 21st to 27th)

  Apr 27, 2026 |  3 minute read

  These are the current online posts that I enjoyed reading and made me think. What I have been reading and thought worth sharing. AI If you are not the model, you are the harness. Long read, but well worth reading to stay on top of the latest thoughts on harness engineering. - link [opinion] - ( Added: 2026-04-27 08:06:01 ) Read to find out how the SOTA model.

  • bookmarks

• What I have been reading this week (March 29th to April 20th)

  Apr 20, 2026 |  9 minute read

  These are the current online posts that I enjoyed reading and made me think. AI Great video that provides an essential overview of how to build with AI (in this instance using Spring AI). Has a supporting Github repo at https://github.com/tzolov/voxxeddays2026-demo - link [tutorial] - ( Added: 2026-04-20 17:14:20 ) Agent harness or harness engineering is the new hotness right now, so this is a breakdown of all the components need to build an ai coding agent - link [opinion] - ( Added: 2026-04-20 10:34:17 )

  • bookmarks

• What I have been reading this week (17th - 28th March)

  Mar 28, 2026 |  3 minute read

  This is what I read about and tried this week. AI Scion is an experimental multi-agent orchestration testbed designed to manage concurrent LLM-based agents running in containers across your local machine and remote clusters. It enables developers to run groups of specialized agents with isolated identities, credentials, and workspaces, allowing for a dynamic and evolving graph of parallel execution of tasks such as research, coding, auditing, and testing. - link [tool] - ( Added: 2026-03-28 12:50:21 )

  • bookmarks

• What I have been reading this week (10th - 16th March)

  Mar 16, 2026 |  4 minute read

  This is what I read about and tried this week. AI Find out how Anthropic models are being used across geographical regions. Broken down by use cases - link [report] - ( Added: 2026-03-15 14:49:37 ) A look at a new tool to sandbox AI agents on MacOS, which the zooms out and looks at some of the broader questions you need to ask yourself - link [tool] - ( Added: 2026-03-15 14:37:21 )

  • bookmarks

• What I have been reading this week (2nd - 9th March)

  Mar 9, 2026 |  4 minute read

  This is what I read about and tried this week. AI A graph-based tool routing library for LLM agents. Instead of asking your LLM “what tool should I use next?” at every step, Self-Healing Router models your tools as a weighted directed graph and uses Dijkstra’s algorithm for deterministic routing. When a tool fails, edge weights update to infinity and the path recomputes automatically. The LLM is only called as a last resort — when no feasible path exists.

  • bookmarks

• What I have been reading this week (1st March)

  Mar 1, 2026 |  7 minute read

  Currently reading ( 1st March) These are the current online posts that I enjoyed reading and made me think. AI AWS responsible use of AI guide (pdf) - link [best-practice] - ( Added: 2026-03-01 15:16:45 ) Agentic AI is an amplifier, for both good and bad organisational behaviours. This post looks at some of the things you might observe for those who have not yet got their house in order - link [opinion] - ( Added: 2026-03-01 13:12:29 )

  • bookmarks

• What I am currently reading (23rd Feb)

  Feb 19, 2026 |  3 minute read

  Currently reading ( 23rd Feb) These are the current online posts that I enjoyed reading and made me think. CODING A tool that provides observability across your Claude Code sessions - when running multi agent swarms this provides some nice info on what they are all up to - link [tool] - ( Added: 2026-02-23 17:47:05 ) Brooks proved in 1975 that adding more programmers to a late project makes it later because communication overhead grows faster than output.

  • bookmarks

• Using aws-vault to manage access to your AWS resources from Kiro CLI

  Feb 19, 2026 |  8 minute read

  As we say at AWS, security is our top priority. This is why I have been spending time thinking about how to manage what our agentic AI tools can do. After publishing a previous post on this topic (Getting Kiro CLI to use short lived AWS credentials), this post takes a look at another approach you can take. Vincent reminded me about an awesome open source tool called aws-vault that helps developers move to using short lived, temporary credentials.

  • kiro
  • agentic ai
  • security
  • aws-vault

• Getting Kiro CLI to use short lived AWS credentials

  Feb 19, 2026 |  10 minute read

  Agentic AI tools like Kiro IDE and Kiro CLI are pretty awesome. You provide the right context, the right prompt, and away they will go. After a period of time, you can now review the output. Early on developers understood the power, but also the potential harm that could be cause (and we began to read stories like this one. Most developer tools now provide capabilities that allow you as a developer to have fine grain control of what they can and more importantly, should never do.

  • kiro
  • timebound-iam
  • agentic ai
  • security

• What I am currently reading (17th Feb)

  Feb 17, 2026 |  3 minute read

  What I am currently reading ( 17th Feb) These are the current online posts that I enjoyed reading and made me think. AI A new CNCF inspired look at the agentic AI space - link [tool] - ( Added: 2026-02-22 23:48:00 ) A detailed 100 page guide from Jeremy Daley on how to build production grade multi tenant systems - link [best-practice] - ( Added: 2026-02-22 21:03:48 )

  • bookmarks