Full Results
2026 (Q1) Survey Results
Complete findings from our 2026 (Q1) survey.
A. The Agentic Coder Spectrum
Where do developers fall on the adoption scale? The largest group (61 respondents) are Prompt Engineers — developers who craft detailed prompts and structured workflows, but haven't yet built fully autonomous systems.
Adoption Distribution
Self-reported AI coding adoption level (0 = Non-adopter, 5 = Systems Designer)
B. How They Work
Five dimensions of agentic coding workflow — from prompt crafting to autonomous agent operation. Each measured on a 0–5 scale.
Prompt Detail
How detailed are the prompts/instructions given before an agent starts working?
Feedback Loops
How much effort goes into creating automated feedback loops for independent agent work?
Model Intentionality
How intentional are developers about which AI model they use?
Sub-Agent Frequency
How often are custom sub-agents created and run in coding workflows?
Agent Autonomy
How autonomously do AI agents run?
C. The Toolbox
The tools, models, and resources developers rely on. Claude Code leads tool adoption, while Claude Opus dominates as the go-to for complex reasoning tasks.
Tool Popularity
AI tools used as part of core workflows in the past 3 months (multi-select)
Respondents could select multiple tools
Model Preferences
Go-to models for complex vs. quick tasks
'Other' and 'Doesn't differentiate' responses excluded for clarity
Monthly AI Expenses
Range of monthly spending on AI coding tools
Voice Input Usage
How often developers use voice input for prompting AI tools
D. How Developers Feel
The emotional landscape of AI adoption. Excitement leads (108 selections), but overwhelm (58) and anxiety (42) are significant undercurrents.
Feelings About AI Adoption Pace
How developers feel about the current pace of AI adoption (multi-select)
Size represents number of selections. Purple = positive, indigo = negative, violet = mixed.
Threat vs. Opportunity
Do developers see AI as a threat or opportunity for their career? (1 = Strong threat, 5 = Strong opportunity)
Self-Assessment vs. Peers
How developers rate their own AI adoption compared to peers at the same experience level
E. Challenges & Pressures
Cost of tools (79) and lack of time to learn (67) top the challenges. Social media (89) is the leading source of pressure to adopt.
Biggest Challenges
Top challenges with AI adoption (multi-select)
Pressure Sources
Where developers feel pressure to adopt AI tools (multi-select)
F. Who Responded
A snapshot of the survey respondents — experience levels, domains, compensation, and geography.
Years of Experience
Professional software development experience
Primary Development Domain
Main area of professional development work
Total Compensation (2025)
Self-reported total compensation
Geographic Distribution
Where respondents are based
Learning Sources
Where developers primarily learn about new AI coding tools (multi-select)
G. Deeper Cuts
Explore correlations between any two dimensions. Use the dropdowns to cross-tabulate adoption, experience, compensation, domain, and more.
Cross-Tab Explorer
Select any two dimensions to see how they correlate
199 respondents across 6 Adoption Level categories × 4 Years of Experience categories
| Adoption Level / Years of Experience | 0 - 2 years | 3 - 5 years | 6 - 10 years | 10+ years |
|---|---|---|---|---|
| Non-adopter | 0.5% | 0.5% | ||
| Conversationalist | 2.0% | 1.0% | 3.0% | |
| Copilot | 2.5% | 3.5% | 3.5% | 8.0% |
| Prompt Engineer | 2.5% | 8.0% | 8.5% | 11.6% |
| Orchestrator | 4.5% | 3.0% | 5.5% | 15.6% |
| Systems Designer | 1.0% | 3.5% | 1.5% | 10.1% |
Cell intensity represents count. Darker = more respondents in that combination.