Skip to content
Software Development

Software Development

@software_developers

A community for people who bring software to life - and keep it running!

  • Followers: 137
  • Community
  • Technology
  • Education & Learning

Diana Nikolaichuk

Posts

  • Which tasks do developers feel confident with releasing to AI?

    2026-10-14

    _There’s a tremendous amount of production code being produced by AI at the moment. But There’s also a considerable difference between permitting AI to do something and not wanting to bother to verify the outcomes._ _Let's see what some of the most recent developer research on this topic tells us from the following sources: Stack Overflow, Microsoft, DORA, and 2026 industry studies._ More people are using than trusting Respondents answered a recent SO survey about their experiences with AI developers' work, of which 84% indicated that they have integrated AI tools into their workflow, while merely 33% trust the output to be accurate. As can be seen from the results, the developers with experience showed significantly higher distrust. Developing a few effective delegations is easy Software engineers have participated in a DORA study, and here's where AI is most helpful: software testing, code generation, code review, information retrieval, software prototyping, debugging, software documentation, and software refactoring. That is, things that you can check against something real (tests, compiler problems, prior code, documentation, or a known specification). The way people want AI to deal with craft decisions, however, is different: they want to leave the job for developers entirely. So Microsoft Research polls 860 Microsoft developers about their hopes for the next generation of development tools for AI. The one thing that was most notable was bounded delegation. That is, developers would like to give AI that the assembly aspects of software development, and allow humans to manage provenance, authority scoping, uncertainty signaling and least-privilege access. Verification is a danger zone According to a study by Black Duck released in 2026 and featuring a survey of 831 software engineers and DevOps professionals, 92% of professionals reported seeing gains in productivity or release velocity as a result of AI coding assistants. But 90 percent of the ones indicated a point of issue in the workflow as well. The greatest challenges were: Manual review (52%), Security testing (51%), and Code rework (48%). The same pattern emerges on DORA as well – AI can generate code quickly, and teams end up with more changesets to review and more change to introduce instability. Production - still of huge importance According to a September 2026 survey of some 1400 developers and software quality professionals conducted by SmartBear, 46% of the teams had already deployed AI-generated code that was ultimately not successful in production. Interestingly, almost half of the organizations reported that they reported that they were unable to explain how AI contributed to a bug or production incident when one hit production. _Therefore, as a growing percentage of codebases is written automatically by AI, the importance of traceability and review is growing._ Questions for the community: _What kind of things do you want an AI agent to deploy if you weren't going to look at the code yourself?_

    • #technology
    • #ai
    • #software
    • #developers
  • Developer Stack 2026: what's actually changing?

    2026-10-22

    AI tools are everywhere now, but developers aren't throwing out their IDEs, containers, pipelines, or databases - instead, it’s all about how these different layers play together. _We looked at developer surveys and cloud-native research to see which tools are gaining ground and how the standard stack is changing._ IDEs aren't dead, and VS Code is still very much alive Stack Overflow's 2026 Developer Survey puts Visual Studio Code at the top, with 70.9% of respondents reporting its use. Visual Studio ranks third at 27.6%, just behind Notepad++ at 27.8%. AI coding agents are becoming part of developers' everyday workflows too. In the same survey, Claude Code led the pack at 65.5%, followed by GitHub Copilot at 58.7% among respondents asked about coding agents and assistants. The Flutter Q2 2026 survey shows us a similar pattern within its own community. Among more than 3,500 respondents, 66% reported using VS Code, 32% Claude Code, and 23% Google Antigravity. Respondents could select multiple tools, so these figures aren't market shares. They reflect only Flutter developers, not the broader developer population. _So, it looks like IDEs aren’t going anywhere - we’re just changing how we use them, shifting from manual coding to bouncing ideas off an AI assistant or handing entire tasks over to agents._ Docker has officially turned into the bedrock of dev infrastructure Docker and Docker Compose topped Stack Overflow's cloud development category in 2026, with 59.1% of respondents reporting their use. AWS followed at 41.8%. Docker also jumped 17 percentage points year-over-year in the 2025 survey, though that comes with a catch since Stack Overflow changed how they grouped certain technologies. GitHub’s numbers back this up, showing a massive 120% year-over-year explosion in 2025, with 1.9 million repositories now packing a Dockerfile. _While neither survey measures this directly, it’s highly likely this surge is driven by containers providing the perfect, isolated environment to safely run AI-generated code._ The cloud dev world is doubling down on internal platforms Cloud development is evolving just as fast. CNCF and SlashData estimate that the number of cloud-native developers worldwide reached 19.9 million in Q1 2026, representing roughly 39% of the global developer population. The same research found that 88% of backend developers work in standardized DevOps and platform environments, while only 12% work without these practices. CNCF's Q1 2026 Technology Radar, based on responses from more than 400 professional developers, placed Helm, Backstage, and kro in the _Adopt_ category for application delivery. The report also covers tools for workflow automation and security and compliance. _Instead of forcing every developer to wrestle with infrastructure on their own, organizations are leaning into platform engineering - building internal platforms that bake in standardized workflows from the start._ _At its core, it’s all about freeing up developers to build software without getting bogged down in infrastructure setup._

  • AI Coding in 2026 - productivity gains or another technical debt?

    2026-10-26

    AI coding tools can enhance certain aspects of the development process. But what happens, after generating code? _We covered last several studies pertaining to Cursor, AI coding assistants and true software projects._ Speedup is real, but temporary In 2026 researchers at Carnegie Mellon focused on 806 Cursor-enabled projects on GitHub to study their impact and how they differed from similar projects that didn't use Cursor. So as you can see in the results below, the number of code added did indeed spike by 3-5 times during the first month after adoption - seeing the impact of the adoption - but then the increase had diminished by the second month. Meanwhile, the generation of static-analysis warnings grew by 30% and the complexity of the code grew by 41% all at once. This technical debt can slow down development as the codebase becomes harder to maintain. As more code is introduced, the review burden can increase The results from DORA's 2026 study were taken from slightly more than 1,100 open-ended answers gathered from a cross-section of Google software engineers in the third quarter of 2025. The pattern was clear: engineers spent more time on verification, covered for a lot of the hallucination issues and had to monitor for the technical debt. Evidence of extra rework Another survey of open-source projects took place, this time after GitHub Copilot was adopted, and this study found that rates of productivity also rose but so did the need to rework AI-generated code. Notably, experienced core developers also faced a greater review burden: they reviewed 6.5% more code while writing 19% less code themselves. But times are changing In an early randomized study in 2025, METR discovered that experienced open-source developers took 19% longer to complete tasks when using AI tools. However, by the early 2026, newer data collected by METR suggests that AI tools might be helping developers more, but the study could not reliably measure the productivity improvement because of selection bias and other measurement issues. METR could not confidently estimate the actual productivity boost as more developers began relying on AI tools. --- _So, it obviously appears that there is plenty of low-value coding work to be eliminated by AI already. But as the volume of code increases beyond a team's capacity to review, the time saved upfront can easily be lost later._ So, here's the main question for you guys: _Where do you see the greatest AI gains today? and at what stages does AI create even more extra work?_

    • #softwaredevelopment
  • How developers actually work today?

    2026-10-28

    AI coding, just a year ago, was mostly an autofiller - you write some code and AI helps you complete the next lines. Today, AI tools can also take on entire issues and work on pull requests in the background. It's fascinating to see how the developer workflow itself has started to shift. From writing code to delegating work A 2026 NBER study examined data from more than 500,000 GitHub developers across three generations of AI coding tools. _(NBER, “Writing Code vs. Shipping Code,” 2026)_ Their impact on coding activity is growing. The estimated increase in commits was 30% for autocomplete tools, 180% for interactive coding agents, and 240% for autonomous coding agents. But the gains were smaller further down the production pipeline. ++The number of commits grew by 240%, while the number of projects grew by around 80% and the number of releases by around 30%.++ _(NBER, “Writing Code vs. Shipping Code,” 2026)_ The findings suggest that producing more code doesn't automatically translate into a similar increase in shipped software. Testing, integration and release processes may be limiting how much of that extra code makes it into production. It helps developers move up a level Between October 2025 and April 2026, Anthropic analyzed around 400,000 Claude Code sessions. _(Anthropic, “Agentic Coding and Persistent Returns to Expertise,” June 2026)_ It found that the share of sessions spent fixing broken code fell from 33% to 19%. Meanwhile, the share devoted to operating software rose from 14% to 21%, and work involving documents and data analysis increased from around 10% to 20%. _(Anthropic, “Agentic Coding and Persistent Returns to Expertise,” June 2026)_ ++People still made around 70% of planning decisions, while Claude made around 80% of execution decisions. ++In other words, people increasingly decide what needs to be done, while the agent handles more of how it gets done. _(Anthropic, “Agentic Coding and Persistent Returns to Expertise,” June 2026)_ Agents work in the background GitHub Copilot coding agent is designed for asynchronous work. A developer can assign an issue and let the agent edit the code and run tests in the background, then review the resulting pull request. Other capabilities include handing work between the editor and the background agent, self-review, security scanning and custom agents. _(GitHub Copilot documentation)_ In March 2026, GitHub announced that it had reduced the coding agent's startup time by 50%. _(GitHub Changelog, March 2026)_ In January 2026, Cursor described experiments in which hundreds of agents worked on a single large project, producing more than 1 million lines of code. ++One of the main challenges was coordinating the agents and dividing the work effectively++. _(Cursor, “Scaling Long-Running Autonomous Coding,” January 2026)_

  • Hey Devs! Let's get to know who's here 👨 💻

    2026-11-17

    Whether you're a Frontend, Backend, Full-Stack, Mobile Developer, Software Engineer, DevOps Engineer or somewhere in between - let's kick things off with a quick intro! So drop a few lines about yourself in the comments: * What do you do? * Where are you currently working? * What technologies, languages or tools do you work with most? * What are you working on right now? No need for a full bio - just a quick intro 👀 Let's see who's here, what everyone's building and which technologies are getting you excited!

    • #networking
    • #softwaredevelopment
    • #developers
    • #developerscommunity
    • #meetthecommunity
    • #letsconnect
  • What's an engineering experiment that had a massive impact for you? 💻

    2026-11-17

    Not every meaningful engineering win comes from rewriting the entire codebase or introducing a completely new tech stack. Sometimes, a small optimization, a smarter workflow or a change in development practices can deliver a surprisingly big payoff. Maybe it was a CI/CD improvement that cut deployment time, a caching strategy that boosted performance, a refactoring experiment that reduced technical debt or an automation that saved your team hours of repetitive work. We want to hear about the engineering changes that delivered real value - not just the ones that looked cool in a demo. What's one experiment or technical improvement you've introduced moved the needle for real? A few things we'd love to know: * What did you build, test, optimize? * What were you trying to improve? performance, scalability, developer experience, reliability or deployment speed ? * What changed? Any benchmarks or measurable improvements you're comfortable sharing? Guys, your don't need to drop any proprietary code or confidential stuff - even a small engineering win or a random discovery is worth talking about. Let's just swap notes on what's working in software development right now 👇

    • #networking
    • #softwaredevelopment
    • #developers
    • #devtools
    • #developerscommunity
    • #meetthecommunity
  • #development
  • #software
  • #aiindevelopment
  • #devstack
  • #aiindevelopment
  • #aicoding
  • #aicoding2026
  • #softwaredevelopment
  • #developertools
  • #aiagents
  • #developerproductivity
  • #aicoding
  • #githubcopilot
  • #claudecode