Perete Harrison

This video features a step-by-step walkthrough of a Google code lab on building distributed multi-agent AI systems, led by Perete Harrison (DevOps Software AI Engineer) and Koye Adeyombo (Front-end Development Engineer) from Atop Web Technologies. Video Overview: Building Multi-Agent AI Microservices [RAW] In this comprehensive session, the engineers demonstrate how to go beyond simple chatbots to create a sophisticated, distributed system using the Agent Development Kit (ADK) and Google’s Gemini models. As this is a raw, unedited walkthrough, viewers will see the engineers encounter and troubleshoot real-world issues, such as credential and project resolution errors, in real-time. Perete Harrison explains the underlying concepts as they navigate the code lab, which was originally created by Amit Maharaj. They build a course creation system where specialized agents—a Researcher, a Judge, and a Content Builder—coordinate through an Orchestrator to produce high-quality educational conten...

  No doubt we have moved on from generation to verification . I mean, for the past few years, the amount of code I write with my own hands has dropped drastically. In some cases, almost to zero. Most times now, I am verifying output, reviewing generated code, checking if it actually solves the problem, catching hallucinations, and maintaining what comes out of the AI. That is a big shift. Years back, loose prototyping, implicit shortcuts, clever ways of doing things, all of that made sense. The developer was the one writing the code, so if it made sense in your head and saved you some time, fine. But now, when a lot of the code is generated for you, the question becomes different. Can I easily understand what this thing is doing? Can I verify it? Can I quickly spot when something is wrong? This is one reason I keep liking Go more. Of course, Go is my favorite programming language, so I may be biased. But seriously, Go code written by a senior engineer, a junior engineer, or generat...

  To me, as I advanced in programming, I came to realize the true difference between programming and software engineering. Software engineering is applying engineering principles to build software web applications, desktop applications, mobile apps, or services. But as you progress, you see it's much more about collaboration. Working with other engineers to architect, design, build, test, implement, and deploy robust, high-performance, predictable systems that can evolve over time. Programming is writing code to solve a specific problem. And that's part of it. But software engineering is the broader discipline. Collaboration and process that make those solutions durable and scalable. The gist is, programming focuses on writing code to solve a problem. Software engineering includes that, and adds design, architecture, testing, deployment, and teamwork to build systems that last.

  To architect, design, build, test, and deploy predictable applications and services, planning is inevitable. Given the pitfalls of coding agents and their non-deterministic nuances, we are excited to announce AXON , an open-source protocol and plugin from Atop Web Technologies for AI coding agents like Codex, Antigravity, and Claude Code. AXON stands for Atop eXecution Orchestration Network . Its main purpose is to solve a huge problem in AI-assisted development: agent drift . When building with AI, agents are very fast, but they often lack durable context between sessions. They forget product intent, design systems, tech stack decisions, workflows, testing expectations, and why certain decisions were made in the first place. That is where unpredictable results start. AXON introduces a protocol to fix this: Context first, specification second, implementation third, verification always. What AXON Does With AXON, the agent does not just jump into code. It initializes the project c...