Forte Group

Your engineers are faster. Your delivery metrics aren't. That's the AI adoption trap most engineering orgs walked into in 2024, and most haven't walked out of yet. We keep seeing the same 5 failure patterns across orgs of every size: 1️⃣ The workflow was never redesigned 2️⃣ Adoption is measured wrong: seat count ≠ impact 3️⃣ Every role needs a different integration pattern 4️⃣ The first win has to be voluntary, not mandated 5️⃣ Nobody owns the change end-to-end The hard truth: you probably have 2026 tools running inside 2023 workflows. Full breakdown + what actually works 👇 https://lnkd.in/etcW9Etd

“Two pizza teams” are becoming obsolete. According to Invisible Technologies' Platform CTO Kit Colbert, the future might look more like: 👉 2–3 engineers 👉 Coordinating dozens of AI agents 👉 Shipping faster than entire teams today On the latest episode of the CTO2CTO podcast, we explore: — Why innovation has nothing to do with “newness” — How vMotion changed the industry (not just the tech) — Why AI won’t replace engineers, but will redefine them The real shift is already happening, from writing code → to orchestrating systems. Watch/listen to the full episode here: https://lnkd.in/efNavbMM

“AI-native” isn’t about adding tools. It’s about rethinking how products are built from the ground up. At Xceptor, that meant moving beyond isolated use cases and embedding intelligent agents across the entire delivery lifecycle, while keeping human oversight where it matters most: governance, trust, and decision-making. The result? Faster delivery, less rework, and a system that scales with confidence. Learn more about turning AI from experimentation into real, operational change: https://lnkd.in/eaqRFmue

A new issue of Forte Spotlight is out! This month, we focus on a shift that’s easy to miss: AI is no longer just supporting systems. It’s becoming part of how they operate. Inside this edition: • A new threat model: the “AI-in-the-middle” attack pattern • Why reasoning is the missing layer in AI-assisted engineering • How APIs are replacing traditional interfaces • What real-world systems teach us about reliability and scale Read the full issue 👇

🎙️A New CTO2CTO episode is live! Lucas Hendrich sits down with Kit Colbert, Platform CTO at Invisible Technologies (and former VMware CTO), for a masterclass in what actually drives innovation. Kit's prediction: By the end of 2026, AI agents will run autonomously for a week. Not 12 hours. A week. We also cover: → Why invention ≠ innovation (lessons from building Storage vMotion) → The shift from two-pizza teams to half-pizza teams → Why developers will cost 2x their salary in token spend → What Kit learned from "totally messing up" as a GM One of our favorite episodes yet. 🎧 Watch/listen now: https://lnkd.in/efNavbMM

What does it take to automate customer communication without losing control over quality and accuracy? In this case study, we break down how an automated email response solution was designed to handle real-world complexity, while keeping outcomes reliable and production-ready. Key takeaways for engineering leaders: → AI-driven automation needs strong validation layers to avoid risk → Quality engineering must evolve to test dynamic, non-deterministic outputs → Observability and feedback loops are critical for continuous improvement Read the case study: https://lnkd.in/eH3kbr-Q

We let AI manage our entire dev backlog: requirements, user stories, Jira tickets, test cases, docs, release notes. One engineer put it plainly: "We spend a surprising amount of time describing software instead of building it." So we tested whether AI agents could change that. Results from the experiment: — User story: 30 min → 10 seconds — Test cases: 60 min → 15 seconds — Release notes: 15 min → 5 seconds The human still reviews everything. The mechanical work disappears. The full write-up covers the 8-agent pipeline we built, how it connects to Jira and Confluence, and what breaks if you skip the instruction-file tuning step. Link to insights👇 https://lnkd.in/exRHPdTe

AI is changing how software is built,but it’s also changing how it needs to be tested. Traditional QE approaches aren’t designed for systems that learn, evolve, and behave unpredictably. That’s why testing AI-enabled systems requires a fundamentally different mindset. Here’s how we approach quality engineering for AI-driven solutions: https://lnkd.in/eK8XCKKR

Everyone is experimenting with AI, but very few teams are actually operationalizing it across software delivery. At CTO Craft, Michael Kinloch (SVP Engineering at Xceptor) shared how they’re moving beyond pilots, working with Forte Group to build an AI-native product development lifecycle. From introducing AI into day-to-day workflows… → to scaling it across the SDLC → to enabling agents to drive parts of delivery (with humans in the loop) The result? Work that used to take weeks or months is now happening in hours. This is what it looks like to move from AI experimentation to real transformation. Watch the full talk here: https://lnkd.in/eaqRFmue

The SaaS playbook is dead. For 15 years, the formula worked: → Great design → Content team → Organic traffic → ARR In 2025, all three collapsed. UI is commoditized: what cost $150K in design now costs $20 in API credits. SEO is evaporating: 69% of searches end without a click. Stack Overflow lost 98% of its questions to AI. So what survives? API gravity. Stripe isn't worth $70B because of its dashboard. It's worth $70B because 500 other tools depend on its API every day. The new moat isn't what users see. It's what other software can't live without. We wrote about how to make that shift, and what to stop doing immediately. 🔗https://lnkd.in/ewkcMQZg