borghei/Claude-Skills / c-level-advisor/vpe-advisor
vpe-advisor
VerifiedVP of Engineering advisor on org design, productivity, quality, delivery, and capacity planning. Use when scoring engineering org health, designing the eng org, planning capacity, or building the productivity dashboard.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 12.1 KB |
| references/eng-productivity-and-quality.md | markdown | 9.1 KB |
| references/eng-strategy-and-roadmap.md | markdown | 7.6 KB |
| references/engineering-org-design.md | markdown | 9.8 KB |
| scripts/eng_capacity_planner.py | script | 11.0 KB |
| scripts/eng_org_health_scorer.py | script | 17.7 KB |
| scripts/eng_productivity_dashboard.py | script | 10.8 KB |
references/engineering-org-design.md
View on GitHub# Engineering Org Design Reference
Practical reference for designing or restructuring an engineering organization
at any stage from 5 to 5000 engineers.
## 1. Org shapes (by stage)
### Stage 1: < 30 engineers — Functional or single team
- One pool of engineers; specialization emerges informally
- Single tech lead; founder often plays VPE/CTO
- Hiring: focus on generalists who can ship end-to-end
- Org risk: bottleneck on the founder/lead
### Stage 2: 30–100 engineers — Squad emergence
- 4–8 product squads of 4–8 engineers
- Squads paired with product manager + designer
- Platform / DevOps team forms (3–6 people)
- Roles: engineering manager (EM) per squad; founding VPE
- Org risk: weak EM bench; squad sprawl
### Stage 3: 100–500 engineers — Multi-squad with platform
- Squads grouped into pillars or domains
- Platform team graduates into platform org (15–30 people)
- Dedicated infrastructure, security, data, ML engineering
- VPE + CTO are clearly separate roles
- Layered EM structure (Director level emerges)
- Org risk: cross-team dependencies; platform becoming blocker
### Stage 4: 500–2000 engineers — Pillar + business unit
- Engineering org mirrors business structure
- Multiple platform sub-teams (FE platform, BE platform, data platform)
- Multiple specialist teams (security, SRE, ML, growth)
- VPs of Engineering per BU; SVPE or CTO over all
- Org risk: standards drift; fragmented eng culture
### Stage 5: 2000+ engineers — Enterprise engineering
- Domain-aligned engineering orgs
- Centralized platforms with strong product mindset
- Dedicated developer experience function
- Multiple CTOs/VPEs across product lines
- Org risk: bureaucracy; reorg-driven attrition
## 2. Common org pitfalls
### "Two-pizza" myth at scale
The classic 6-person squad doesn't scale beyond ~50 engineers without
a coordination layer. At 100+ engineers, the *coordination structure*
matters more than squad size.
### Platform as a service catalog with no UX
Many platform teams ship tools that engineers technically can use but
don't want to. Treat the platform as a product with internal users.
### Specialists in product squads
Embedded specialists (e.g., ML engineer, security engineer) work when
the squad has steady demand. Otherwise, they become an expensive
generalist. Pool them centrally and assign on-demand.
### Re-org as a productivity strategy
Re-orgs cost ~1–2 quarters of velocity. Use them to fix structural
problems, not to look like you're doing something.
## 3. Roles and titles
### Individual contributor (IC) track
- **Sr Engineer** (~3-5y) — owns features end-to-end; mentors juniors
- **Staff Engineer** (~6-10y) — sets technical direction for a domain; mentors seniors; partners with EMs
- **Sr Staff Engineer** (~10y+) — technical direction across multiple domains
- **Principal Engineer** — organization-wide impact; sets engineering-wide standards
- **Distinguished / Fellow** — exceptional cross-org influence; external presence
### Manager track
- **Engineering Manager (EM)** — manages 5-8 engineers; pairs with product / design
- **Sr EM** — manages 8-12 engineers; or 2 EMs
- **Director** — manages 3-5 EMs (~25-50 engineers)
- **Sr Director** — manages multiple Directors
- **VP** — owns a pillar / org (~50-200 engineers)
- **SVP / CTO / VPE** — owns the engineering function
### Career ladder principles
- IC and EM tracks should be equal in pay and prestige
- IC track must offer growth without forcing management
- Promotion criteria must be transparent and consistent
- Calibration sessions across managers prevent drift
## 4. Hiring sequence by stage
### Stage 1 — < 30 engineers
- Hire generalists (full-stack, can deploy infra)
- Hire for ramp-up speed and judgment
- Avoid early specialists (ML, security, SRE) unless core to the product
### Stage 2 — 30–100 engineers
- Hire founding VPE if founder hasn't earned the role
- Add EMs as squad count grows past 4
- First specialist hires: SRE, security, data engineer
- Begin formal interview training
### Stage 3 — 100–500 engineers
- Distinct CTO + VPE
- First Director of Engineering hire
- Specialist functions: ML, growth, mobile, infra
- Dedicated recruiting + sourcing function
- Begin platform investment
### Stage 4 — 500–2000 engineers
- VPs of Engineering per pillar / BU
- Heads of: SRE, Security, Data Engineering, ML Engineering, Platform, DevEx
- Engineering Operations / Chief of Staff role
- Performance management / promotion calibration becomes formal
### Stage 5 — 2000+ engineers
- Multiple SVPs / VPEs
- Distinguished / Fellow engineers as recognized senior IC
- Dedicated developer relations
- Internal engineering brand + external
## 5. Squad design principles
### Size: 4–8 engineers
- Below 4: bus risk; on-call unsustainable
- Above 8: communication overhead grows quadratically
- Sweet spot is 5–7
### Cross-functional pairing
- Squad pairs with PM + designer (when product-facing)
- Platform squads may not need PM but need a clear customer
### Mission clarity
- Each squad has a 1-page mission: what they own, what they don't,
who they serve, success metrics
### Stable membership
- Squads form for a quarter+ minimum; rotation hurts more than it helps
- Exception: rotation rotations to platform / pairing assignments
### On-call ownership
- Squad owns on-call for its services
- 6–8 engineers minimum for sustainable rotation
- Shared on-call across multiple squads as fallback
## 6. Platform engineering — what to build
A modern platform team owns the "developer paved path":
- CI/CD pipeline (templates, opinions, fast)
- Service templates with built-in observability + security
- Infrastructure abstractions (Kubernetes, serverless, where appropriate)
- Internal developer portal (Backstage-style)
- Secret management
- Observability stack (metrics, logs, traces)
- Security tooling (SAST, dependency scanning, SBOM)
- Cost attribution
- Deployment safety (feature flags, canaries — see `engineering/feature-flags-architect`)
A successful platform team:
- Measures adoption (% of services on the paved path)
- Treats internal engineers as customers (NPS, friction interviews)
- Has a product manager (yes, even for internal tools)
- Is staffed at ~10–15% of total engineering
## 7. The Engineering Manager role
A good EM does five things, in priority order:
1. **People** — 1-on-1s, career growth, conflict, hiring, performance
2. **Delivery** — squad is shipping the right things on the right cadence
3. **Quality** — technical decisions are sound; on-call is healthy
4. **Process** — squad rituals work; not too heavy, not too light
5. **Strategy** — squad mission is clear; partners with PM on prioritization
EMs that focus only on delivery become project managers. EMs that focus
only on people become career coaches. The good EMs do both.
### EM-to-engineer ratios
- New EM: 4–6 reports
- Experienced EM: 6–8 reports
- > 10 reports = drowning; split the team
### EM promotion criteria
- Demonstrated coaching outcomes (engineers promoted, regrettable retention)
- Squad delivery health (DORA, on-call)
- Hiring outcomes (offers accepted, ramp time)
- Org influence (improving processes beyond own squad)
## 8. Performance management
### Cadence
- 1-on-1: weekly
- Formal check-in: monthly or quarterly
- Performance review: semi-annual or annual
- Calibration: at each review cycle
### Performance distribution
- ~10–15% top performers (strong promote / extra reward)
- ~70–80% solid contributors (continued investment)
- ~5–10% improvement needed (PIP or restructured scope)
- Force-ranking is poisonous; use distributions as a check, not a target
### PIP (Performance Improvement Plan) discipline
- PIPs are real; people pass them when given the chance
- Document expectations; weekly check-ins; clear endpoint
- If PIP is a euphemism for "we want them out," skip it and offer severance
## 9. Engineering culture — what actually matters
Culture is built through:
- **Hiring bar** — who you hire and reject
- **Promotion criteria** — what gets rewarded
- **Code review norms** — kindness + rigor; not gatekeeping
- **Incident response** — blameless postmortems with action items
- **On-call** — equitable, supported, with rotation discipline
- **Decision-making** — who decides, on what basis, who knows
- **Documentation** — what's written down vs tribal
Culture is NOT built through:
- Slogans
- Team-building exercises
- Perks
- Conference talks
- Carving "what we believe" into a wall
## 10. Diversity, equity, inclusion
A mature engineering org takes DEI seriously, operationally:
- Diverse interview panels
- Calibrated rubrics for hiring + promotion
- Inclusive on-call (don't penalize parents / different time zones)
- Sponsorship programs that go beyond mentorship
- Comp transparency (at minimum: bands by level)
- Demographic data tracked and reviewed (with privacy care)
- Inclusion is in scope for managers' performance reviews
## 11. Distributed / remote / hybrid
### Fully remote
- Async-first culture
- Documentation is mandatory
- Synchronous time is for collaboration, not status
- Geographic diversity is a strength
### Hybrid
- Specify in-office days clearly
- Design for hybrid meetings (camera + audio quality)
- Manage proximity bias in promotion
### In-office
- Increasingly rare past 2024
- Justify with specifics, not "culture"
## 12. Common pitfalls
- **EM track without IC track.** Forces strong engineers into management.
- **One-size-fits-all squad model.** Different teams need different shapes.
- **Re-org as productivity intervention.** Costs more than it saves.
- **Platform team without product manager.** Builds what engineers want, not what they need.
- **No on-call rotation discipline.** Burns out top engineers.
- **Specialist roles embedded too early.** Underutilized; expensive.
- **Hiring bar lowered in growth pushes.** Quality debt compounds.
- **Skipping calibration.** Promotion drift; comp inequity.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |