The context
Before making good developer tools, we needed a point of view on where software development was actually heading and who our developers were becoming. I built a structured breakdown of how AI adoption is reshaping developer behaviour across three distinct personas and presented to C-suite as a strategic input for prioritization.
Results at a glance
C-suite
Presentation delivered
Future of SDLC analysis
3
Developer personas mapped
Legacy · Augmented · Autonomous
1
Strategic direction set
CEO aligned on next target segment
The friction
As AI tools reshaped how developers worked, the team had no shared answer to a critical question: who exactly is our developer, and how is that changing? Individual squads were making IDE and product decisions without a common view of how developer behaviour was actually evolving. This meant we risked building for a developer that no longer existed, or ignoring the ones that were emerging.
What I found
Through desk research, tool benchmarking, workflow modelling, and synthesis with PM and engineering leads, I identified three distinct developer personas emerging from AI adoption:
- Legacy DevX: Skeptical developers who believe in coding as a craft. They use AI as a replacement for search, nothing more. Every suggestion is checked. Sonar's role here is familiar: traditional scan and fix in an existing workflow.
- Augmented DevX: Developers using AI for isolated tasks, troubleshooting, and generation to move faster. This is the dominant pattern today. Sonar’s presence in the pre-PR environment needs to be stronger and seamless.
- Autonomous DevX: Skilled prompt engineers who offload the majority of implementation to agents and act primarily as code reviewers and validators. For this persona, Sonar's role shifts fundamentally from helping humans write better code to becoming the trust and governance layer for agent-produced code.
The key non-obvious finding: Sonar will have customers from all three categories simultaneously. The strategic question isn't which persona to serve. It's how to provide the right experience for each, and how to ensure that as developers move along the spectrum.
The decisions this enabled
- Gave the team a concrete, shared vocabulary for discussing who their developers were and how that was changing — replacing vague references to "the developer" with three specific, distinct user models.
- When presented to C-suite, the analysis resonated immediately. The CEO's direct response: Sonar should prioritize building for developers where AI adoption is already high (Augmented and Autonomous segments) as that's where the market is heading and where the competitive window is narrowest.
- Informed the squad's product direction across all developer tooling — IDE extension, CLI, and MCP. The framework gave every squad member a shared lens for evaluating what to build next and for whom, regardless of which surface they were working on.
Risks and tradeoffs
- Vision without validation: The three personas are directionally right. But they're built from desk research and benchmarking, not direct user validation. The risk is that the adoption spectrum reflects how experts think about AI, not how the median Sonar customer is actually experiencing it. Focused discovery to validate adoption personas and enterprise trust thresholds is the next step.
- Speed of change: The SDLC landscape is moving fast enough that a persona framework built today may need significant revision in 12-18 months. The risk isn't being wrong - it's ensuring we put in consistent effort to ensure personas are valid as trends change.
What stayed human — and why
Deciding which developer segment to prioritize first wasn't a data call. It was a judgment call about competitive urgency, Sonar's existing strengths, and where the market window was narrowest. The analysis narrowed the space and gave leadership a shared frame. The prioritization decision required human conviction, not just evidence.
Result: Future of SDLC analysis delivered to C-suite · Three developer personas defined · CEO aligned on prioritising Augmented and Autonomous DevX as next strategic target
If I ran this again…
- The entire analysis had to be delivered in one week so there was no time for primary research or customer validation. The personas were built from desk research, benchmarking, and synthesis with internal experts. If I had more runway, I'd validate the adoption spectrum directly with customers at different points on it, which would sharpen the boundaries and surface needs the desk research inevitably missed.