Developer ramp-up time continues to accelerate with AI
Engineering Enablement · email · 2026-04-09 Reader 🖥️
AI-assisted developers are reaching their 10th merged PR in roughly half the time compared to pre-AI baselines, with the trend continuing to accelerate quarter-over-quarter, though this metric captures speed-to-milestone without measuring the quality or depth of understanding behind those contributions.
theoretical · expository — A short newsletter reporting empirical findings on developer onboarding speed. It presents data and interprets a trend, but does not argue for a specific action. The implicit practical implication ('use AI to accelerate onboarding') is never stated as a recommendation.
Structure
| Part | Role |
|---|---|
| Historical context and Q4 2025 baseline | Establishes prior evidence -- the Q4 2025 finding that daily AI users reached 10th PR in 49 days vs. 91 days for non-AI users |
| Q1 2026 update with methodology | Core evidence -- presents the new data point (33 days as of April 2026, down from 39 in Q4 2025) from a 400-company sample of 500+ developer organizations |
| Interpretation of the trend and diminishing returns question | Analytical framing -- asks whether the trend will continue or hit a natural floor, noting the shift from individual AI adoption to organizational mandates |
| Limitations acknowledgment | Self-critique -- concedes that Time to 10th PR does not capture quality, rework, or depth of understanding |
Key terms
- Time to 10th PR Measuring AI Impact
The average number of calendar days between a new developer's start date and their 10th merged pull request, used as a proxy for onboarding ramp-up speed - ramp-up time Measuring AI Impact
The period from hire to productive contribution, operationalized specifically as Time to 10th PR rather than any broader notion of competence or autonomy - daily AI use Measuring AI Impact
Engineers who show daily usage of AI tools (the specific measurement method is not disclosed -- could be tool telemetry, survey self-report, or activity logs)
Propositions
- explicitAverage Time to 10th PR has dropped from ~70+ days in Q1 2024 to 33 days in April 2026, a reduction of more than 50%.conf: high
- explicitThe quarter-over-quarter improvement from Q4 2025 to Q1 2026 was approximately 15% (39 days to 33 days), smaller than the earlier 50% drop but still meaningful.conf: high
- explicitPart of the recent improvement comes from AI shifting from optional individual adoption to organizationally mandated or guided use during onboarding.conf: high
- explicitcorroboratedTime to 10th PR does not measure the quality of contributions, rework generated, or depth of understanding of new hires.conf: highsupports ←Guide to scaling engineering organizations · p8 — This piece's /dev/start outcome data (90% stack understanding via a 3-4 week mentored project) is the pre-AI depth-of-understanding baseline that extract-developer-ramp-up-accelerates-ai's own gap (p4: time-to-10th-PR doesn't measure depth of understanding) identifies as missing from its AI-accelerated ramp-up data.
- explicitThere may be a natural floor to onboarding time compression where remaining steps are not easily compressed by AI.conf: medium
- implicitMerged PRs are a meaningful proxy for productive contribution during onboarding.conf: high
- implicitFaster ramp-up is desirable and valuable, independent of whether faster-ramped developers produce equivalent long-term outcomes.conf: high
- implicitThe AI tools used in onboarding primarily compress the 'figuring things out' phase -- understanding codebases, architecture, and conventions -- rather than the coding itself.conf: medium
Arguments
AI is meaningfully accelerating developer onboarding as measured by time to first productivity milestones. strength: medium
Explicit premises:
- In Q4 2025, daily AI users reached 10th PR in 49 days vs. 91 days for non-users.
- In Q1 2026, the average dropped further to 33 days.
- The dataset spans 400 companies with 500+ developers each.
Implicit premises:
- Time to 10th PR is a valid and sufficient proxy for onboarding effectiveness.
- The sample of 400 companies is representative of the broader industry.
- The 'daily AI use' cohort is comparable to earlier cohorts in all relevant ways except AI usage.
- The definition and measurement of 'daily AI use' has remained consistent across quarters.
Inferences:
- If the trend continues but with diminishing returns, there is a lower bound set by irreducible non-AI onboarding activities (social integration, access provisioning, domain knowledge that is not codified).
- The shift from individual to organizational AI adoption may be a one-time step-function improvement rather than a compounding trend.
The organizational mandate to use AI during onboarding is partly responsible for the continued improvement. strength: low
Explicit premises:
- AI is no longer optional in many organizations.
- New hires are guided to use AI for codebase questions, architectural understanding, and draft changes.
Implicit premises:
- Organizational mandates increase actual usage, not just reported usage.
- The onboarding AI use cases (asking questions, understanding architecture) are genuinely effective, not just displacing other effective learning methods (e.g., pairing with a senior engineer).
Inferences:
- If the improvement is partly a selection effect (organizations that mandate AI are also organizations with better onboarding infrastructure in general), the AI-specific contribution is overstated.
Narratives
- The progression from Q1 2024 baseline (~70+ days) through Q4 2025 (49 days for AI users) to Q1 2026 (33 days) as a multi-quarter trend line
→ Creates a momentum narrative -- the trend is continuing, inviting the reader to project further improvement and invest in AI-enabled onboarding
Questions addressed
- ✅ Is AI continuing to accelerate developer onboarding, or was the Q4 2025 finding a one-time effect?
The trend is continuing: Time to 10th PR dropped another 15% QoQ from 39 to 33 days. - ⭕ Is there a natural floor to how much AI can compress onboarding time?
Explicitly flagged as an open question. The diminishing rate of improvement (50% then 15%) may suggest convergence, but agentic solutions could push it further. - ⭕ Does faster Time to 10th PR mean new hires are actually onboarded effectively?
Acknowledged as unresolved. Quality, rework, and depth of understanding are not measured. Promised for future analysis. - ⭕ What is driving the continued improvement -- individual AI skill or organizational enablement?
The article suggests organizational mandates play a role but does not quantify the relative contribution.
Gaps
- incompleteThe article measures speed to a PR milestone but never addresses whether AI-accelerated onboarding produces developers who understand systems as deeply as those who onboarded more slowly. Faster PRs could mean shallower understanding, higher downstream rework, or increased dependence on AI as a crutch rather than developing genuine codebase knowledge. The article acknowledges this gap but treats it as a future research item rather than a fundamental limitation of the current finding.
Read next: Look for research on knowledge retention and long-term developer effectiveness when onboarding is AI-assisted vs. traditional. The Dreyfus skill acquisition model and research on desirable difficulties in learning suggest that compressing the 'figuring things out' phase may have hidden costs. - incompleteThe cross-sectional comparison from Q4 2025 (49 days with AI vs. 91 days without) suffers from the exact selection bias DX/Abi Noda has warned about in other publications: engineers who adopt AI daily may be systematically different from those who do not. The Q1 2026 data shifts to a longitudinal comparison (same cohort, AI users, tracked over time) which is methodologically better, but the headline '50% reduction' still rests on the weaker cross-sectional design.
Read next: DX's own meta-longitudinal study (referenced in the companion podcast extraction); academic literature on selection effects in technology adoption studies. - incompleteNo discussion of what 'daily AI use' means operationally. Is it measured via tool telemetry (Copilot usage logs), self-report, or organizational policy? The definition matters enormously for interpreting the results -- if it is policy-based ('we tell everyone to use AI'), it conflates organizational onboarding quality with AI-specific effects.
Read next: DX methodology documentation; compare with how other studies define and measure AI adoption intensity. - uninformedThe article does not consider that the PR bar itself may be shifting. If organizations are breaking work into smaller PRs (a common practice enabled by AI-assisted coding), then 10th PR arrives sooner not because the developer is ramping up faster but because the unit of work has shrunk. Without controlling for PR size or complexity, the metric is ambiguous.
Read next: Research on PR granularity trends in AI-assisted development; DX's 'true throughput' metric which weights by lines of code.