A report released today reveals a disconnect between engineers and decision makers and that only one in four teams report smooth workflows as AI adoption, brittle systems, and tooling misalignment drive backend complexity.

The survey of over 220 software engineers, architects, infrastructure contributors, and engineering leaders worldwide comes from Temporal Technologies, creators of the Durable Execution platform. Conducted in Q2 2025, it looked into the tools, workflows, and challenges shaping development in modern infrastructure environments in order to report on the current state of development in software engineering. 

According to Samar Abbas, co-founder and CEO of Temporal Technologies:

"The report tells us a lot of what we hear from partners every day – backend challenges aren’t just technical, they’re also organizational, Engineers and decision makers are prioritizing different things, and that disconnect is driving tooling delays, reliability risks, and rising complexity across the stack. AI is only adding another layer of scale and unpredictability.”

The report leads on AI telling us:

Everyone is playing with AI, but few are actually building for it. A whopping 94% of professionals say they’re using AI in their workflows (mostly through tools like Copilot or ChatGPT), but only 39% are working on building reliable frameworks to support it at scale. That’s like buying a Ferrari and parking it in a garage made of popsicle sticks.

Asked about challenges to workflow management only a quarter of respondents reported there were none and there were some obvious differences between individual contributors (software engineers) and decision makers:

The top concern of high operational overhead was reported by 35% of respondents - 38% of decision makers but only 32% of individual contributors.The gap was even more extreme in the case of difficulty scaling which was a concern for 35% of decision makers but only 26% of individual contributors.

The top concerns for individual contributors were difficulty in managing long-running processes and complex failure recovery, reported by 36% and 34% respectively, while both were only reported by 26% of decision makers. 

The report argues that the lack of common ground is at the root of the problems felt by teams:

The shared pain here is more cultural than technical. What’s missing is trust. ICs are in the trenches, patching outages and jury-rigging solutions. DMs are managing budgets, compliance risk, and exec expectations. Both sides are exhausted, but they’re rarely aligned on what the actual problems are, let alone the solutions.

Having identified the problem, the report also suggests the answer - to use tooling to bridge the gap. It suggests that tools can serve: 

as a common language that can give everyone a shared language on the same page and transform alignment from a pipe dream into actionable progress.

Moving on, the report looks at tools used by large (more than 1000 employees)  and small (less than 1000 employees) organizations:

The main finding is that workflow orchestration tools are the most widely adopted solution overall, used by 53%, followed by scheduled jobs, 45%. In both cases they are used more by large companies than small ones. However it is custom solutions that show the biggest gap between small and large companies. Nearly half of large companies (49%) rely on custom-built systems, compared to just a third of smaller ones, with the report commenting:

big companies have the budget to build their own problems.

Larger teams are also more likely to lean on task queues (45%) and streaming platforms (31%), while smaller teams keep things simpler, relying more heavily on serverless setups and event-driven architecture.

The report also notes:

More than a third of professionals at companies with over 1,000 employees juggle three or more tools to manage their workflows. In smaller orgs, only 26% reach that level of complexity. The takeaway? As companies scale, they tend to accumulate tools like a software hoarder.

The report concludes with a plug for Temporal's Durable Execution, which allows developers to focus on business logic rather than infrastructure complexities, increasing developer velocity and is designed to allows you to write application code that is crash-proof and resilient to failures. An open source product it is free if self-hosted or paid-for as a managed service, Temporal Cloud.

More insights are included in the State of Development Report 2025.

More Information

State of Development Report 2025

www.temporal.io

Related Articles

Stack Overflow Reveals Trends In AI Adoption

JetBrains Reports On Software Quality

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: [email protected]