The start of a new year always invites predictions. At RedMonk we don’t approach making predictions as though we’re making bets. There’s Polymarket for that (unfortunately.) Instead, we find the act of articulating what we think might plausibly happen next to be useful because it forces us to more fully examine the present.  

We view making predictions as an exercise in trying to sort signal from noise. It pushes us to surface our assumptions and pressure-test our internal narratives before we externalize them. In that sense, predictions are less about seeing the future and are more about seeing the present more clearly. Importantly, however, we’ll caveat this batch of predictions with our usual disclaimer, which is that often we can tell you what will happen, we just can’t tell you when.  

With that framing, I asked each RedMonk analyst to offer a single prediction for 2026. These aren’t meant to be exhaustive, consensus-driven, or even mutually reinforcing. Instead, they reflect some of our technical, organizational, economic, and cultural vantage points across the developer ecosystem.  

Enjoy, and happy new year!

- Rachel and The Monks

Subscribe now

RedMonk 2026 Predictions

Kate: 2026 is the year web developers rediscover design as a differentiator. The proliferation of AI-generated, shadcn-fabulous websites has created a sea of ho hum sameness. Developers and companies are recognizing that standout design makes visitors stop, look, and actually feel good about a product. When everyone can spin up a competent-looking site with AI, the websites that truly invest in craft and visual experience will be the ones that stand out.

Rachel: The deadline for the Cyber Resilience Act (CRA) is approaching quickly, with potentially far-reaching consequences for organizations selling software into the European market. As the first horizontal regulation of the software industry, the CRA represents a fundamental shift that the industry is largely unprepared for. I expect the back half of 2026 to be marked by a fair amount of chaos as organizations grapple with compliance, not just at a technical level but, in some cases, through significant adjustments needed in to their go-to-market strategies.

Steve: with 2025 cementing the role of agents in the coding assistance space, it is gradually shifting a subset of developers’ primary focus from that of pure builder to more of a supervisory architect type role. As such, two non-exclusive approaches to improving both development velocity and code quality. First, single agent repetitive refinement (SARR, or just ARR if you like pirates), in which a given coding agent operates repetitively on the same asset, continuing to refine and improve it rather than relying on one shot approaches. Second, multi-agent swarms, in which developers increasingly unleash so-called coding swarms to work in effect as virtual software teams. The latter approach, in turn, will create a management and orchestration problem to be solved, much as Kubernetes was developed to manage fleets of containers.

James: Post-quantum cryptography (PQC) is going to be a thing in 2026. You've likely heard some of the hype about quantum computers breaking cryptography. We're not going to be seeing quantum in general production this year, next year, or even before 2030. But you need to address PRQ right now. Maybe not if you're a start-up or even a scale-up but if you're an enterprise the EU roadmap for PQC asks member states to start transitioning by the end of 2026, and to secure high-risk systems by the end of 2030, using standardized post-quantum cryptography algorithms. Talk to auditors of European banks and compliance organisations and they’re all over this. In the USA - the National Security Agency ’s Commercial National Security Algorithm Suite 2.0. is still looking at time frames to be ready in the early 2030s. In pattern this is somewhat like the Y2K bug: we’re going to need to evaluate, build an inventory of systems, applications and exposures, and then mitigate. That’s across all legacy code and production systems.

Links Roundup

• As someone who used to chart their commute time to try to isolate the impacts of route and time of departure, this analogy sings to Rachel.

• While the actual motivations have been periodically disputed over the years, one of the more common justifications for Google’s decision to release Kubernetes as open source software was as a strategic way to provide a layer of insulation between workload and underlying cloud platform, such that apps that went into AWS would not be essentially welded to that platform. A decade plus later, Bryan Cantrill makes the argument here that that effort was successful.

Recent RedMonk Research

• In the afterglow of 2025’s avalanche of AI developer tooling announcements, Kate Holterhoff is revisiting the “10 Things Developers Want from AI Code Assistants” series she authored in 2023 and 2024 for RedMonk because THIS YEAR THINGS HAVE REALLY CHANGED. RM clients mentioned: AWS, Microsoft/GitHub, Red Hat, Tessl, and Google

• What do VMs, microservices and agents have in common? And what does Yeats have to do with all of that: The Blood Dimmed Tide of Agents

Recent RedMonk Videos

• In this RedMonk conversation, Mike Hartington, Developer Advocate at Prisma, chats about mobile development with Kate Holterhoff. They discuss mobile frameworks, the challenges of cross-platform development, and why CI/CD is so hard for deploying mobile apps: Mike Hartington on the State of Mobile Development & Why CI/CD is Still So Hard

Recent Media Appearances

• James Governor tells how decades of software delivery practices are now essential for managing AI’s unpredictable behavior, in this episode of The New Stack Makers: Why You Can’t Build AI Without Progressive Delivery

• Beyond the P99 moments are the rare, unpredictable outliers that disproportionately affect performance, reliability, and user experience. In this session, Rachel Stephens sits down with engineering leader Adrian Cockcroft to explore what really happens past the 99th percentile. From his early work on embedded systems to chaos engineering at Netflix to building modern AI-assisted analytics, Adrian shares decades of lessons on how to uncover, understand, and fix the long tail of system behavior: Performance Insights Beyond P99: Tales from the Long Tail

Sponsor Monki Gras!

Monki Gras, the intimate London conference about software, craft, and tech culture, returns on March 19th & 20th, 2026.

This year’s theme is Prepping Craft – being prepared in software and life.

In an era where uncertainty is the only certainty, we will explore the strategies that make people, communities, products, and platforms safer and more resilient. We’ll take a meta view, linking software testing strategies and resilience engineering directly to real-world preparedness, covering everything from managing cloud outages to living off-grid.

Interested in sponsoring the conference? Please reach out to Morgan Harris for options!

Monki Gras Tickets are still available…

Early bird options are sold out, but regular priced tickets are available here!

Meet the Monks

Events we'll be attending:
- Dynatrace Perform: 26 - 28 January 2026, Las Vegas
- Kubecon EU 2026: 23 - 26 March 2026, Amsterdam
- SUSECON: 20 - 21 April 2026, Prague

Events we'll be hosting:
- Monki Gras 2026: 19 - 20 March 2026, London, UK