# rafal-wenvision-ingenierie-logicielle-ere-ia-tout-change-rien-ne-change-2026-06-01

## Veille

Op-ed by **Olivier Rafal** (Consulting Director Strategy, **WeNvision** — **SFEIR** group; former editor-in-chief of *Le Monde Informatique*) published on **June 1, 2026** on **CIO-Online**, structured around a **paradox**: in the age of AI, software engineering **changes everything… and nothing changes**. **What changes is the operating model.** Roles are redefined: the **Product Owner** shifts from backlog breakdown to **generating context usable by AI**; the **developer** shifts from writing code to **framing, steering, and reviewing** agent execution; **QA** gains the ability to define **expected proof** upfront. Team structure shifts from *"double pizza teams"* (hand-off chains of ~8 people) to ***"sandwich teams"***: a **tight pairing of a business expert and a tech lead, both AI-augmented**, with other skills in support. Internal **Sfeir** figure: *"this pair now drives roughly 80% of the production chain,"* the remaining ~20% (architecture, data governance, security) being centralized. Pivot quote: ***"The issue isn't a tooling issue, it's an operating-model issue."*** **What doesn't change is the discipline of the cycle.** The **SDLC** phases (define → build → verify → deploy → maintain) remain identical and non-negotiable; AI removes none of them, it **intensifies** them: ***"all the slack that human pacing used to absorb, however imperfectly, becomes, at AI speed, industrial-grade defects"*** (an amateur-vs-professional sports metaphor). Hence **three inviolable *gates*** (human control): **specification, planning, delivery review**; validation **by proof** (not by AI's own assertions); **systematic capitalization** (each cycle feeds the next) → measured result: **−30% correction iterations after ~10 cycles**. Principle: ***"the faster the execution, the stricter the framework must be."*** Concepts invoked: **harness** (agentic rules adapted to context), **vibe-coding** judged **untenable in the enterprise**. **Third pillar = governance, FinOps & value-driven management**: **variable and recurring** AI costs (~**€10/hour** per augmented role), shift from flat-fee licensing to usage-based billing (a parallel with cloud in the 2010s); **FinOps** does not aim to cut costs but to *"optimize tool efficiency"* (cost relative to value); aligning **business metrics** upfront (time-to-market, features, performance, eco-design). **Conclusion**: acceleration makes the fundamentals **non-negotiable**; the challenge is **organizational and cultural**, not technological — without securing the business relationship and collective discipline, an AI-powered SDLC only **amplifies the problems** (driving into the wall faster). Extends the WeNvision doctrine of [[rafal-wenvision-ia-generative-produit-techno-pas-projet-2024-02-23]] and [[rafal-wenvision-tokenomics-foundation-finops-ia-2026-06-04]]; converges with *systems around the model* dropbox-okumura-beyond-code-generation-engineering-productivity-ai-agents-2026-05-28, *harness engineering* osmani-agent-harness-engineering-2026-04-19, agentic Salesforce, and the *agent manager* debate (BFM/Girard, SFEIR).

## Titre Article

L'ingénierie logicielle à l'ère de l'IA : tout change... et rien ne change

## Date

2026-06-01

## URL

https://www.cio-online.com/actualites/lire-l-ingenierie-logicielle-a-l-ere-de-l-ia-tout-change-et-rien-ne-change-17062.html

## Keywords

software engineering, AI, everything changes nothing changes, operating model, Target Operating Model, TOM, redefined roles, Product Owner, context generation, developer framing steering review, QA expected proof, double pizza teams, sandwich teams, business expert tech lead pair, 80% production chain, Sfeir, SFEIR, centralized architecture security data, SDLC, define build verify deploy maintain, cycle discipline, industrial-grade defects, AI speed, amateur professional sport, gates, specification planning delivery review, validation by proof, capitalization, -30% correction iterations, 10 cycles, the faster the execution the stricter the framework must be, harness, harness, vibe-coding untenable, governance, FinOps, FinOps applied to AI, 10 euros per hour augmented role, variable recurring costs, usage-based billing, cloud parallel, optimize efficiency, value-driven management, time-to-market, eco-design, organizational and cultural challenge, amplify the problems, Olivier Rafal, WeNvision, CIO-Online

## Authors

**Olivier Rafal**, *Consulting Director Strategy* chez **WeNvision** (groupe **SFEIR**). Ancien **rédacteur en chef du *Monde Informatique***, et auparavant consultant analyste du marché IT (~10 ans). Tribune publiée dans la rubrique *Tribune* de **CIO-Online**. Publié le **1er juin 2026**.

## Ton

**Profile**: Senior consultant-expert op-ed (CIO *thought leadership*), addressed to **CIOs and technical leadership**, **analytical-prescriptive** register, moderate-to-high technical level (SDLC, operating model, FinOps, harness). Dual authority stance: consulting (WeNvision/SFEIR) + former reference IT journalist.

**Style**: Built as a **diptych** ("everything changes" / "nothing changes"), then a **triptych** with a 3rd pillar (governance/FinOps/value). Deliberate use of **internal Sfeir figures** (80%, −30%, €10/h) that lend credibility, **metaphors** (amateur vs. pro sport), and punchy **signature formulas**. Anti-hype and anti-tool-solutionism: insists the stakes are **organizational and cultural**. Dense, operational text, oriented toward CIO decision-making.

**Key aphorisms**:
- ***"Everything changes… and nothing changes."*** (paradox-title).
- ***"The issue isn't a tooling issue, it's an operating-model issue."***
- ***"All the slack that human pacing used to absorb, however imperfectly, becomes, at AI speed, industrial-grade defects."***
- ***"The faster the execution, the stricter the framework must be."***
- (FinOps) *"optimize tool efficiency"* (≠ reducing token consumption).

**Metaphors / frameworks used**:
- ***Everything-changes / nothing-changes paradox*** — the operating model reinvents itself, the SDLC cycle stays invariant but demands more rigor.
- ***Amateur vs. professional sport*** — at AI speed, small lapses become disqualifying.
- ***Sandwich teams*** — a business+tech lead pair, AI-augmented, driving ~80% of production (vs. double pizza teams).
- ***The 3 gates*** — specification / planning / delivery review as inviolable human checkpoints.
- ***Harness*** — contextualized agentic rules framing execution.

**Epistemic position**: opinion op-ed **backed by a quantified internal track record** (Sfeir), which sets it apart from a pure editorial. Explicit consulting purpose (WeNvision). Authority reinforced by the author's background (former editor-in-chief *Le Monde Informatique*). To be read as a credible **transformation framework**, with internal proof points taken as indicative.

**Authority**: (a) **dual legitimacy** — consulting + IT journalism; (b) **Sfeir field figures** (80% / −30% / €10/h); (c) **structured framework** (diptych + 3 pillars + 3 gates), directly actionable; (d) **doctrinal consistency** with his 2024-2026 op-eds.

## Pense-betes

- **Date / source**: **June 1, 2026**, **CIO-Online** (Op-ed). Author: **Olivier Rafal**, Consulting Director Strategy **WeNvision (SFEIR group)**, former editor-in-chief *Le Monde Informatique*.
- **Paradox thesis**: ***"everything changes… and nothing changes"*** — the **operating model** reinvents itself, the **SDLC cycle** stays invariant but becomes **non-negotiable**. ### What CHANGES (operating model)
- **Roles**: PO → **generates context** usable by AI; dev → **frames/steers/reviews** agent execution; QA → defines **expected proof** upfront.
- **Teams**: *double pizza teams* (≈8, hand-offs) → ***sandwich teams*** (a **business expert + tech lead** pair, AI-augmented, other skills in support).
- **Sfeir**: this pair drives *"roughly 80% of the production chain"*; ~20% (architecture, data governance, security) **centralized**.
- Quote: ***"the issue isn't a tooling issue, it's an operating-model issue."*** ### What DOESN'T CHANGE (discipline of the cycle)
- **Invariant SDLC**: define → build → verify → deploy → maintain; AI **intensifies**, removes nothing.
- *"The slack […] becomes, at AI speed, industrial-grade defects"* (amateur/pro sports metaphor).
- **3 inviolable gates**: **specification / planning / delivery review**.
- **Validation by proof** (not AI's own assertions) + **capitalization** → **−30% correction iterations after ~10 cycles**.
- ***"The faster the execution, the stricter the framework must be."*** Concepts: **harness**, **vibe-coding untenable** in the enterprise. ### 3rd pillar: governance / FinOps / value
- **Variable & recurring** AI costs ≈ **€10/hour** per augmented role; flat fee → **usage-based** (parallel with cloud in the 2010s).
- **FinOps ≠ cost-cutting**: *"optimize tool efficiency"* (cost/value), not just token consumption.
- Align business metrics **upfront**: time-to-market, features, performance, **eco-design**. ### To use in engagements / presentations
- **Ready-to-use transformation framework** (diptych + 3 pillars + 3 gates) — particularly relevant on the **SFEIR/consulting** side (internal proof points).
- Connects to: *systems around the model* (Dropbox/Okumura), *harness engineering* (Osmani), agentic Salesforce, token FinOps (his own Tokenomics Foundation op-ed), *agent manager* doctrine (BFM/Girard).
- Reusable formula: *"the faster the execution, the stricter the framework must be"* = an antidote to vibe-coding and to *"driving into the wall faster."*

## RésuméDe400mots

In this op-ed from **June 1, 2026** on **CIO-Online**, **Olivier Rafal** (Consulting Director Strategy at **WeNvision**, **SFEIR** group, former editor-in-chief of *Le Monde Informatique*) argues a paradox: in the age of AI, software engineering **changes everything… and nothing changes**.

**What changes is the operating model.** Roles are being redefined: the Product Owner shifts from backlog breakdown to **generating context usable by AI**; the developer moves away from writing code to **frame, steer, and review** agent execution; QA can define **expected proof** upfront. Structures evolve from *"double pizza teams"* (eight-person hand-off chains) to ***"sandwich teams"***: a tight pairing of a business expert and a tech lead, both AI-augmented, with other skills in support. At Sfeir, *"this pair now drives roughly 80% of the production chain,"* the remaining 20% (architecture, data governance, security) being centralized. The formula sums it up: *"the issue isn't a tooling issue, it's an operating-model issue."*

**What doesn't change is the discipline of the cycle.** The SDLC phases — define, build, verify, deploy, maintain — remain identical; AI abolishes none of them, it **intensifies** them. *"All the slack that human pacing used to absorb, however imperfectly, becomes, at AI speed, industrial-grade defects"* — like sport, whether amateur or professional. Hence **three inviolable *gates*** (specification, planning, delivery review), **validation by proof**, and systematic **capitalization** that cuts **correction iterations by 30% after about ten cycles**. Guiding principle: *"the faster the execution, the stricter the framework must be."* The **harness** frames the agents; **vibe-coding** is judged untenable in the enterprise.

**Third pillar: governance, FinOps, and value-driven management.** AI costs are variable and recurring (~€10/hour per augmented role); as with cloud in the 2010s, the shift is from flat fee to usage-based billing. FinOps does not aim to cut costs but to *"optimize tool efficiency,"* relating cost to value (time-to-market, features, performance, eco-design).

Conclusion: acceleration makes the fundamentals non-negotiable; the challenge is first and foremost **organizational and cultural**, not technological. Without a healthy business relationship or collective discipline, an AI-powered SDLC only amplifies the problems.

## GrapheDeConnaissance

- Olivier Rafal —publie→ L'ingénierie logicielle à l'ère de l'IA : tout change... et rien ne change (DOCUMENT, 0.97)
- Olivier Rafal —travaille_chez→ WeNvision (ORGANISATION, 0.95)
- WeNvision —fait_partie_de→ SFEIR (ORGANISATION, 0.92)
- Olivier Rafal —affirme_que→ l'enjeu est le modèle opérationnel, pas l'outil (AFFIRMATION, 0.96)
- sandwich teams —remplace→ double pizza teams (CONCEPT, 0.92)
- Olivier Rafal —mesure→ le binôme métier + tech lead pilote ~80% de la chaîne de production chez Sfeir (MESURE, 0.9)
- Olivier Rafal —affirme_que→ l'IA n'abolit aucune phase du SDLC, elle les intensifie (AFFIRMATION, 0.95)
- Olivier Rafal —affirme_que→ « tous ces relâchements que le rythme humain absorbait tant bien que mal deviennent, à la vitesse de l'IA, des défauts industriels » (CITATION, 0.93)
- Olivier Rafal —recommande→ trois gates inviolables (spécification, planification, revue) (METHODOLOGIE, 0.93)
- capitalisation —réduit→ de 30% les itérations de correction après ~10 cycles (CONCEPT, 0.9)
- Olivier Rafal —affirme_que→ « plus l'exécution est rapide, plus le cadre doit être strict » (CITATION, 0.95)
- Olivier Rafal —affirme_que→ le vibe-coding est intenable en entreprise (AFFIRMATION, 0.88)
- Olivier Rafal —mesure→ coût IA ~10 € par heure par poste augmenté (MESURE, 0.88)
- FinOps appliqué à l'IA —améliore→ l'efficience des outils (coût rapporté à la valeur), plus que la réduction des coûts (CONCEPT, 0.92)
- réussite de l'ingénierie IA —est_basé_sur→ un défi organisationnel et culturel (CONCEPT, 0.92)

---
Canonical: https://www.thekb.eu/en/fiches/rafal-wenvision-ingenierie-logicielle-ere-ia-tout-change-rien-ne-change-2026-06-01/