Talsco Weekly: 🏗️ Your IBM i AI Roadmap: Build or Buy?

Welcome to another edition of Talsco Weekly

IBM i Brief: 🔗 IBM just gave AI agents a direct line to IBM i. 👥 COMMON named its new board. 🎯 IBM Think 2026 had one message: focus.
AI: 🔵 IBM i is built for the AI era. 🔧 Build or Buy? The IBM i AI Question Every Shop Is Facing.
Career: 📓 Five entries a day. 200 days. Career changed.
Development: 🖨️ Turn IBM i spool files into PDFs in under 30 seconds. 🧩 Parsing IBM i command list parameters in RPG doesn’t have to be a mystery.
Hiring: 🤝AI is changing what IBM i shops need to hire for.
Leadership: 🏢 Where your IBM i team works is now a talent strategy decision.
Modernization: 📘 The IBM i modernization book that should have existed. 🔄 IBM i modernization is no longer a question of if, but how fast.
Trends: 📈 Eight years of Talsco Weekly tells the story of AI’s rise on IBM i.

IBM i Brief

🔗 IBM just gave AI agents a direct line to IBM i

IBM has released an open-source MCP (Model Context Protocol) server for IBM i, letting AI tools like Claude, VS Code Copilot, and Bob query Db2 for i databases using natural language.

Built on the Mapepire WebSocket gateway, it supports performance monitoring, security auditing, job management, and more through simple YAML-defined SQL tools. Your IBM i data, now AI-accessible.

Looking for documentation on IBM i MCP Server? You can find it here.

👥 COMMON named its new board

For those who have served on any kind of board, you know it’s a tremendous commitment of time, and the community is stronger because people step up to do it.

With that in mind, COMMON just announced its newly appointed board and officers.

The slate includes:

For the rest of the COMMON Board, you can find them here.

🎯 IBM Think 2026 had one message: focus

Constellation Research CEO Ray Wang recapped the event with a clear takeaway: IBM is going deep, not wide. AI, hybrid cloud, and quantum are no longer emerging bets; they are core enterprise strategy.

IBM’s pitch centers on a full-stack blueprint from infrastructure to orchestration, paired with long-term partnerships like Saudi Aramco and Cleveland Clinic. Strategy by subtraction. Know where to double down.

AI

🔵 IBM i is built for the AI era

At POWERUp 2026, IBM made its clearest case yet that IBM i’s integrated architecture positions it ahead of open-systems competitors scrambling to build data lakes.

If you want to run an AI model with production enterprise applications, you would be hard pressed to find a more perfect system than IBM i.

IBM Bob is expanding, MCP server support is live, and agentic AI workflows are emerging. The pitch: stop preparing for AI and start running it.

AI on Power: IBM’s goal is to run AI as close to the data as possible, keeping enterprise AI workloads resident on Power, with Spyre accelerators unlocking larger models and higher throughput.
Agentic Native OS: IBM i already has the five core capabilities agentic systems require: scheduling, identity, state management, auditability, and security. Out of the box. No bolt-ons needed.
The New Backend: Yesterday’s frontends are today’s backends. AI agents now drive IBM i the same way human users once did, making the platform more relevant, not less.

🔧 Build or Buy? The IBM i AI Question Every Shop Is Facing

Someone has been quietly blazing a trail inside a heavily customized IBM i environment, connecting modern AI tooling to a real-world production system. What started as internal experimentation became an unexpected case study in one of the biggest decisions IBM i shops will face.

Does this sound familiar?

We run a highly customized IBM i environment: ERP, EDI, reporting, workflows, built up over years. It works. It runs the business.

Where This Journey Started

This isn’t a developer experimenting on the side. This is a VP of Information Systems modernizing manufacturing operations through systems, automation, and AI.

The kind of architect who knows the IBM i inside out and has worked across enough technology stacks to build something real.

That combination is rare.

When he started experimenting with LLMs, he didn’t write a think piece. He built something, used it in production, learned from it, and shared it.

This comes at just the right time.

There are two types of IBM i shops we typically come across. The first groupfeels stuck. AI is accelerating fast enough that it’s almost paralyzing. So where do you even start?

And then there’s a second group: leaders and decision-makers who didn’t grow up on IBM i. They inherited the platform. They struggle to wrap their head around it, aren’t sure what questions to ask, and their perceptions of what IBM i is or isn’t capable of are often wrong. That misread sends shops down costly paths before a single informed conversation has taken place.

This article is a useful antidote to both.

The build vs. buy question wasn’t the original point, but it’s what surfaces when you read carefully. And it’s the question every IBM i shop will face.

There is no universal answer. A highly customized environment with internal bandwidth looks very different from a standard ERP shop that needs vendor support and governance.

We’ve covered IBM Bob going GA and the tooling race arriving on IBM i. Both paths are real.

The best part: he open-sourced it.

That instinct to share rather than keep it close is exactly what has made this platform and COMMUNITY resilient for decades.

If you’re feeling stuck or trying to make sense of where IBM i fits in your AI roadmap, this is a must-read.

Career

📓 Five entries a day. 200 days. Career changed

Commit to this, and see what happens.

Development

🖨️ Turn IBM i spool files into PDFs in under 30 seconds

Turn IBM i spool files into PDFs using Access Client Solutions, no CL commands or third-party tools required. With AFP Transform Services running on the system, ACS handles the conversion natively. Select the spool file, hit Actions > Download, and drag it to your desktop. Clean, formatted, and email-ready. A practical how-to for developers and admins who still live in the output queue.

For all the details, watch the video 😉

🧩 Parsing IBM i command list parameters in RPG doesn’t have to be a mystery

When building a command that accepts multiple values, the RPG Command Processing Program receives them as a single character string. The first two bytes hold a binary decimal count of entries passed. A data structure maps that count alongside an array overlay of the individual values, letting the program loop only through what was actually entered. A clean, concise pattern every RPG developer should have in their toolkit.

Hiring

🤝 AI is changing what IBM i shops need to hire for

The role of the IBM i developer is shifting. Shops that are modernizing and experimenting with AI tooling need people who can bridge the platform and modern technology stacks.

That profile is rare, and the competition for it is growing.

If your hiring requirements are still written for the developer you had five years ago, you may be looking for the wrong person.

Let’s talk about what that role looks like today.

Leadership

🏢 Where your IBM i team works is now a talent strategy decision

The remote versus office debate has moved past preference. For IBM i shops, it carries real recruiting consequences. Remote-friendly environments unlock a national, sometimes global, talent pool in a field where skilled developers are already scarce. For shops where on-site culture is embedded in the work and the organization, hybrid is the practical middle ground.

Our Take: Neither is wrong. But defaulting to a policy without thinking through talent implications is. Know what you are, own it, and hire accordingly.

Modernization

📘 The IBM i modernization book that should have existed

One practitioner spent 18 months writing the enterprise framework he wished he had: 40 chapters, 4 case studies, 7 named frameworks, roughly 1,000 pages.

The core argument: IBM i is not the problem. The Perception Gap is. Most modernization programs fail because they’re solving the wrong problem.

Want to learn more?

A free Chapter 1 preview is available now with the full launch coming soon.

🔄 IBM i modernization is no longer a question of if, but how fast

The mindset has shifted. At COMMON POWERUp 2026, Integrative Systems reported that more than half of IBM i enterprise leaders they engaged with are actively planning to modernize. The question is no longer whether to move. It’s how to move without breaking what’s working.

Over 50% of IBM i leaders plan modernization within next 12–18 months, with AI accelerating transformation timelines by up to 50%.

The opportunity. IBM i teams that move deliberately, with the right framework and talent, don’t have to choose between stability and progress. They can have both.

Trends

📈 Eight years of Talsco Weekly tells the story of AI’s rise on IBM i

When we launched Talsco Weekly (TW) in 2017, AI didn’t appear once.

By 2018, it showed up a handful of times as IBM corporate news.

For the next four years, it stayed on the periphery. Then ChatGPT hit in late 2022, and the 2023 issues mark the pivot: AI earned its own section header.

In 2024, it was in most issues.

In 2025, nearly all of them.

In 2026, every single one.

Eight years of issues. One unmistakable trend. The IBM i community didn’t miss AI. It waited for the right moment, and that moment is now.

Join

Contact us

If you are an RPG programmer looking to explore opportunities or a client who is looking for a talented IBM i professional, please contact us. We look forward to assisting you.

Do you know of someone who could benefit from Talsco Weekly? If so, please use the social media buttons to spread the word. Thank you!