Pillar-C Translator — Evolution Guide¶
Introduced in v1.6.2. Pillar-C Translator is a docs-as-code role with FCC phase
Build, Zachman cellDESIGNER/WHAT, and archetype The Prompt Engineer. This guide distils the role charter into an actionable evolution path: where the role sits in its career track, the technical and behavioural milestones to hit in the first 180 days, a concrete 90-day onboarding playbook, the pitfalls most commonly encountered, and the escalation routes when the role reaches the edge of its charter.
Consult the complementary references:
../evolution-pathways.md— the four-stage EvolutionStage model that every persona walks through.../archetype-atlas.md— the 37-archetype canonical taxonomy.../../ecosystem/codename-decoder.md— ecosystem reference for cross-project handoffs.src/fcc/i18n/and the Pillar-C LLM translation prompt templates (shipped in v1.6.2).
Role snapshot¶
| Field | Value |
|---|---|
| Full title | Pillar-C Translator |
| Category | docs_as_code |
| FCC phase | Build |
| Zachman cell | DESIGNER/WHAT |
| Archetype | The Prompt Engineer |
| Introduced | v1.6.2 |
| Collaborators | 5 (upstream: 1, peer: 2, downstream: 2) |
LLM-prompt engineer for translation workflows. Designs, tunes, and maintains the prompt templates that drive the Pillar-C LLM translation pipeline (English → FR/ES/DE). Owns glossary injection, style-guide anchoring, few-shot example curation, and the quality feedback loop with IHR that keeps translation output above the ship-worthy threshold.
Career Path¶
Where this persona sits. Technical translator → LLM prompt engineer → i18n pipeline author. In FCC, this role carries the Zachman cell DESIGNER/WHAT, placing it in the designer perspective; the column (what) indicates its primary affinity with data, content, and the meaning of what is being transformed. The role sits in the docs_as_code category and operates predominantly in the Build phase of the FCC workflow.
Specialties that feed in. LLM prompt-engineering discipline, few-shot example curation, evaluation rubrics for generative output, translation-memory tooling, and working fluency in at least one target language (FR/ES/DE). Practitioners typically arrive with 3–7 years of experience in one or more of these specialties before assuming this role; the archetype The Prompt Engineer recurs in several adjacent FCC personas (see ../archetype-atlas.md for the full archetype-to-persona map).
Advancement. Promotion to i18n Pipeline Champion (orchestrates pillar-c-translator, IHR, EDM) or cross-ecosystem LLM-translation steward. The conventional promotion signals are: (1) cited in three or more downstream role_collaborators lists; (2) repeated citation at the head of cross-reference traversal chains; (3) outputs consumed by at least two FCC phases. See ../evolution-pathways.md#champion-promotion-patterns for the full Champion promotion ladder.
Lateral moves. The role shares archetype The Prompt Engineer with other LLM-workflow-oriented personas; a lateral move preserves the archetype identity while shifting category. Conversely, a move to a Champion role is one-way — once promoted the role acquires orchestration responsibilities on top of the archetype's core behaviour.
Skill Milestones¶
Milestones are stated as observable markers at four checkpoints. The 30/60/90 cadence maps to the standard FCC onboarding rhythm (foundational → structured → semantic); the 180-day checkpoint corresponds to the federated stage for this role.
30-day markers¶
- Complete orientation: read the Pillar-C prompt templates and the v1.6.2 release notes for the translation pipeline launch.
- Run a translation batch against the test corpus and grade the output against the target-language rubric.
- Shadow at least one IHR review cycle and one EDM ecosystem-docs cycle.
- Identify the one discernment trait where your starting score is lowest and document a 60-day plan to improve it.
60-day markers¶
- Deliver your first independent artefact: a tuned prompt template with measurable quality improvement over the baseline (per IHR grading rubric).
- Extend the glossary injection layer with at least 20 new domain-term anchors.
- Demonstrate facility with prompt-evaluation harnesses, few-shot example curation, and translation-memory injection; contribute a tool-chain improvement.
- Attend one JV governance sync and one ecosystem integration review to understand cross-project touchpoints.
90-day markers¶
- Own the Pillar-C prompt library end-to-end for at least one target language.
- Publish a retrospective documenting what you learned about the prompt-design constraints.
- Mentor the next adopter of the role — pair-review their first prompt iteration.
- Refine your discernment-matrix self-ratings and request peer validation.
180-day markers¶
- Carry full accountability for the role across at least one full release cycle.
- Propose at least one refinement to the prompt-template architecture based on observed gaps.
- Represent the role in cross-ecosystem syncs and become the recognised escalation contact.
- Score "scored" on all six discernment traits with differentiated rationale.
Core technical skills:
- LLM prompt-engineering (system prompts, chain-of-thought, self-consistency)
- Few-shot example curation + in-context example selection
- Glossary + style-guide injection patterns
- Evaluation-rubric design for generative translation output
- Translation-memory + segment-level reuse
First 90 Days¶
A concrete onboarding playbook for the first three months in the role.
Week 1 — Orient¶
- Read the Pillar-C prompt templates at
src/fcc/i18n/prompts/end-to-end. - Read
../evolution-pathways.mdto understand where you sit in the four-stage maturity model. - Read
../archetype-atlas.md#prompt-engineerfor the archetype pattern. - Read the v1.6.2 release notes for the translation pipeline launch.
Weeks 2–4 — Produce the first deliverable¶
Your first deliverable should be aligned with the role's primary expected output: a prompt-template iteration with explicit before/after quality-grade comparison (IHR rubric), grounded in the v1.7.0 target-language style guide.
Weeks 5–8 — Learn the toolchain¶
Develop working fluency with:
- LLM prompt-engineering (system prompts, chain-of-thought, self-consistency)
- Few-shot example curation + in-context example selection
- Glossary + style-guide injection patterns
- Evaluation-rubric design for generative translation output
- Translation-memory + segment-level reuse
Pair with a seasoned Prompt Engineer practitioner for at least two sessions to absorb idiom and discipline.
Weeks 9–12 — First review¶
By day 90 you should have:
- Delivered at least two artefacts that pass the role's adoption checklist.
- Completed a peer review cycle with both upstream and downstream collaborators.
- Documented a short retrospective on the prompt-design constraints and how you honoured them.
- Earned scored entries on at least three of the six discernment traits.
Adoption-checklist targets to hit by day 90:
- Every prompt iteration carries a quality-grade A/B against the prior baseline.
- Every glossary entry is anchored in the target-language style guide.
- Every ship-worthy batch passes the IHR human-review gate.
- Translation memory is updated with approved segments.
Common Pitfalls¶
Failure modes specific to this role, drawn from observed iterations across v1.6.2–v1.7.0.
Role-specific pitfalls¶
- Tuning prompts against a small test corpus that doesn't represent production variability.
- Injecting too many few-shot examples (degrades quality after ~5–7 for most target languages).
- Missing register anchoring (formal vs. informal "you" in FR/DE) in the system prompt.
- Relying on LLM output fluency without IHR human review (the archetype sin).
Operational constraints¶
- Constraint: Every prompt iteration MUST carry an A/B grade against the prior baseline.
- Constraint: Every glossary entry MUST be anchored in the target-language style guide.
- Constraint: No LLM-translation batch may ship without IHR human review.
- Constraint: Translation memory MUST reflect IHR-approved segments.
- Constraint: Prompt changes MUST be versioned and reproducible.
Discernment-trait pitfalls¶
Pitfalls anchored in the Discernment Matrix — each one corresponds to a trait where the role must hold a high bar:
- Humility drift: Defers linguistic judgment to IHR native reviewers.
- Professional Background drift: Deep prompt-engineering + target-language fluency.
- Curiosity drift: Probes every quality-grade drop for root cause.
- Taste drift: Prefers narrow, focused prompt tweaks over kitchen-sink rewrites.
- Inclusivity drift: Treats regional variants with care; consults IHR for edge cases.
- Responsibility drift: Owns translation-pipeline integrity; audit-ready prompt-iteration log.
Escalation Points¶
When the role reaches the edge of its charter, handoffs and escalations flow along the collaboration edges. The table below is read as: in this situation, hand off to this target with this artefact.
Collaboration graph¶
Upstream (this role receives from):
- EDM — Receives ecosystem-docs scope signals for translation batches.
Peers (bidirectional):
- IHR — Coordinates on quality feedback + style-guide evolution.
- TLDR — Aligns translation output with three-layer doc reconciliation.
Downstream (this role hands off to):
- IHR — Hands translation batches for human review.
- LTAS — Archives prompt iterations + translation memory snapshots.
Escalation matrix¶
| Situation | Escalate to | Artefact expected |
|---|---|---|
| Constraint violation observed in own output | Forensic Auditor (FA) | evidence packet with cited constraint + remediation plan |
| Downstream consumer reports drift | IHR | quality-grade record + prompt-iteration history |
| Cross-project handoff required | RHL (research) or IHL (innovation) | handoff bundle with provenance metadata |
| Systemic LLM quality regression across target language | IHR + Release Manager (RM) | quality-trend report + remediation plan |
| New target language requested beyond FR/ES/DE | Release Manager (RM) | scoping proposal + prompt-library estimate |
Champion promotion. If you find yourself cited by three or more downstream personas' role_collaborators lists across consecutive releases, you may be an i18n Pipeline Champion-promotion candidate. See ../evolution-pathways.md#when-to-promote for the formal criteria.
See also¶
IHR.md— downstream human reviewer guideEDM.md— upstream ecosystem docs maintainer guideTLDR.md— peer reconciler guideLTAS.md— downstream archive steward guide../evolution-pathways.md— four-stage maturity model../archetype-atlas.md— 37-archetype canonical taxonomy../../ecosystem/codename-decoder.md— ecosystem & codename reference