# reock-dx-leadership-ai-engineering-metrics-2025-11-23

## Veille

DX Platform - Engineering Leadership - Productivity Metrics - Psychological Safety - SDLC Constraints

## Titre Article

Effective Leadership in AI-Enhanced Organizations

## Date

2025-11-23

## URL

https://www.youtube.com/live/cMSprbJ95jg?si=4HnxK8w1ELvSr4tz&t=27241

## Keywords

DX, Engineering Leadership, AI Metrics, Psychological Safety, SDLC, Theory of Constraints, Developer Experience, Change Failure Rate

## Authors

Justin Reock (Deputy CTO, DX)

## Ton

**Profile:** Human-Leadership | Systems-Analytical | Educational | Compassionate

Justin Reock adopts a tone focused on thought leadership. He relies on data (Dora, DX data) to deconstruct received wisdom (AI always increases productivity). He places heavy emphasis on the human dimension (psychological safety, fear of replacement) and the systemic dimension (theory of constraints). This is a discourse for managers who want to succeed at AI adoption without burning out their teams.

## Pense-betes

- **Impact Volatility**: Averages hide reality. While the average shows a slight gain, company-level data shows extreme volatility (+20% to -20% confidence). Some companies see their "Change Failure Rate" increase significantly.
- **The Importance of Psychological Safety**: Projet Aristotle (Google). AI is frightening (replacement). Top-down mandates don't work. Reassurance is needed: AI is here to augment, not replace (AI still fails on 2/3 of complex SWE-bench tasks).
- **Measuring Impact (DX Framework)**:
- **Telemetry** (what happens) vs **Survey/Self-reported** (what people feel/context).
- Key metrics: Change Failure Rate, Change Confidence, Maintainability.
- Do not rely solely on usage (copilot adoption).
- **Theory of Constraints**: Saving time on coding is useless if the bottleneck is elsewhere (meetings, waiting for specs, deployment).
- Examples: Morgan Stanley (Legacy code reverse engineering), Zapier (Accelerated onboarding), Spotify (Incident context).
- **Leadership Advice**:
- Unblock usage (Secure sandbox).
- Feedback loop on System Prompts (rule guardian).
- Temperature awareness (creativity vs determinism).

## RésuméDe400mots

Justin Reock, Deputy CTO at DX, addresses the challenge of leadership in AI-adopting organizations. He begins by deconstructing the industry's reassuring averages: while overall slight productivity gains are visible, granular company-level data reveals **extreme volatility**. Some organizations see their Change Failure Rate increase and their developers' confidence collapse.

To succeed, Reock emphasizes **psychological safety** (referencing Google's Projet Aristotle). AI generates fear (of replacement). Leaders must clearly communicate that the goal is capability augmentation, not headcount reduction, especially since agents still fail on the majority of complex autonomous tasks. "Top-down" mandates are counterproductive.

He proposes a measurement framework balancing **telemetry** (what happens technically) and **qualitative data** (developer sentiment), since 95% of productivity depends on the system, not the individual. He warns against the obsession with "time saved coding." Citing the **Theory of Constraints** (Goldratt), he notes that saving an hour on a task that is not the bottleneck is useless.

He cites examples of companies that targeted the right bottlenecks:
- **Morgan Stanley** uses AI to reverse-engineer legacy code (Cobol), unlocking modernization.
- **Zapier** uses bots for onboarding, making new engineers productive in 2 weeks (vs 90 days).
- **Spotify** accelerates incident resolution by automatically pushing context to SREs.

Finally, he offers tactical advice for leaders: establish feedback loops on "System Prompts" (so that AI rules are maintained like code), understand parameters such as "temperature" (creativity vs determinism), and above all, provide secure spaces (sandboxes) so teams can experiment without fear.

## GrapheDeConnaissance

- Justin Reock —travaille_chez→ DX (ORGANISATION, 0.98)
- Justin Reock —affirme_que→ les moyennes cachent une volatilité extrême de l'impact IA (AFFIRMATION, 0.95)
- sécurité psychologique —permet→ adoption IA réussie (CONCEPT, 0.93)
- Projet Aristotle —soutient→ importance sécurité psychologique (CONCEPT, 0.92)
- Google —a_créé→ Projet Aristotle (EVENEMENT, 0.95)
- théorie des contraintes —prédit→ les gains de temps hors goulot d'étranglement sont inutiles (AFFIRMATION, 0.9)
- Morgan Stanley —utilise→ IA pour rétro-ingénierie code legacy (TECHNOLOGIE, 0.88)
- Zapier —utilise→ IA pour onboarding accéléré (TECHNOLOGIE, 0.88)
- Spotify —utilise→ IA pour résolution incidents (TECHNOLOGIE, 0.88)
- mandats top-down —s_oppose_à→ adoption IA (CONCEPT, 0.87)
- hausse du Change Failure Rate —observé_dans→ certaines entreprises adoptant l'IA (CONCEPT, 0.85)

---
Canonical: https://www.thekb.eu/en/fiches/reock-dx-leadership-ai-engineering-metrics-2025-11-23/