🥏 RAI. What is it

We increased our clients’ IT team capacity by 30% with AI for the first time. It only took us 3 years to get to this point 😀 

Here are the details:

Imagine working with a highly knowledgeable team member who can’t focus, has mild dementia, and exhibits sporadic insanity.

No, it’s not a physiological portrait of either of our presidential candidates. It’s an analogy we’ve used five times in the last week in conversations with clients, as it brilliantly explains the head-on approach to using AI for IT organizations.

Off-the-shelf tools (like “Copilot”) can be helpful, but they often produce incorrect or almost correct results—not exactly what we intended.

So the question then becomes:

How can an enterprise IT organization utilize LLM technology to see real RAI (Return on AI—yes, I coined this term) within a few months?

The answer is what we now call Pipe/ines—a needle-moving system of company-specific AI agents.

Each agent can:

•     Create: Code, requirements, tests, etc.

•     Review: Code improvements, suggestions for requirements, missing test cases, etc.

•     Invoke external services: API, build tools, etc.

As a software consulting company, we’re literally disrupting our own successful business model with this approach.

In a few years, it will almost completely replace the need for many engineers we’re being safely paid for today. But if we don’t do it ourselves, the industry will brutally overtake us, and no one will remember our glorious names.

The Pipe/ines setup takes 2-3 months, with immediate results right after.
If you run an IT team that struggles to deliver, is always shorthanded, and feels like every project is a mountain to climb, this is for you.

That’s all for this week…but one more thing. If you’re enjoying this, can you do me a favor and forward it to a friend? Thanks.

-Alex

My incredibly capable software consultancy → Techery

Let’s connect on Twitter/X: @pshenianykov

Or on LinkedIn: Alex Pshenianykov

🫠 THIS MADE ME SMILE