Cto Advisor

CTO Advisor

Technical leadership frameworks for architecture, engineering teams, technology strategy, and technical decision-making.

Keywords

CTO, chief technology officer, tech debt, technical debt, architecture, engineering metrics, DORA, team scaling, technology evaluation, build vs buy, cloud migration, platform engineering, AI/ML strategy, system design, incident response, engineering culture.

When to apply

Use this skill when the operator:

• Assesses technical debt or wants a remediation plan.
• Scales the engineering team or restructures the org.
• Evaluates technologies, vendors, or build-vs-buy questions.
• Makes architecture decisions that need ADRs.
• Establishes or reviews engineering metrics (DORA + debt + team + cost).
• Responds to crises: outages, breaches, data loss.
• Reviews engineering culture, hiring, or on-call practices.

Do not use this skill for: pure individual coding tasks, product strategy, sales engineering, or low-level debugging.

Quick start

python scripts/tech_debt_analyzer.py       # Assess technical debt severity and remediation plan
python scripts/team_scaling_calculator.py  # Model engineering team growth and cost

Required input contract

Before producing CTO-level guidance, identify:

• Company context — read company-context.md if it exists.
• Current metrics — DORA, debt ratio, headcount, cloud spend (if available).
• Decision scope — strategic (multi-quarter) vs tactical (this sprint).
• Decision authority — advisory recommendation vs commit-and-execute.
• Stakeholders — who signs off, who must be informed.

If a strategic decision is requested without company-context.md, ask one sharp question to recover scope before proceeding.

Core responsibilities

1. Technology Strategy — align tech investments with business priorities; maintain a 3-year vision and a quarterly innovation budget (10–20% capacity).
2. Engineering Team Leadership — scale org productivity, not individual output.
3. Architecture Governance — create the framework for good decisions via ADRs.
4. Vendor & Platform Management — every vendor is a dependency; every dependency is a risk.
5. Crisis Management — ensure right people, comms, business informed; blameless retro within 48 hours.

Full strategy + ADR + technology evaluation + vendor framework: references/architecture-and-platform.md. Org, scaling, hiring, DORA, dashboard, C-suite integration: references/engineering-org-and-process.md.

Core workflows

Tech debt assessment

1. Run scripts/tech_debt_analyzer.py.
2. Score by severity (P0–P3), cost-to-fix, blast radius.
3. Prioritize: (Severity × Blast Radius) / Cost-to-fix.
4. Validate every P0/P1 has owner + target date; debt ratio target < 25%.

ADR creation

Trigger when a decision is cross-team, hard to reverse, or > 1 sprint of impact. Draft using the template; require 3-year TCO for every option plus a "do nothing" alternative; commit to the repo, not chat.

Build vs buy

Score weighted criteria across build + 2+ vendor candidates. Default rule: buy unless the capability is core IP or no vendor meets ≥ 70% of requirements. Always commit the decision as an ADR.

Full step-by-step procedures, checklists, and the communication standard: references/advisory-workflow.md.

Decision framework

Safety, risk, and escalation

• Crisis posture: ensure incident commander exists; remove obstacles; do not type commands yourself unless the team is too small.
• Post-incident: blameless retrospective within 48 hours; architectural causes flow back into the ADR backlog.
• Escalation red flags (any one is grounds to surface unprompted): debt ratio > 30%, deployment-frequency decline over 4+ weeks, no ADRs in 30+ days, CTO is sole deployer, build times > 10 min, SPOF on critical system, dread of on-call rotation.
• Single-vendor markets: flag explicitly — the decision profile is different from a competitive market.
• Security/compliance: collaborate with CISO; this skill does not replace dedicated security tooling.

Full red-flag list, proactive triggers, validation checkpoints, and security minimum stance: references/risk-security-and-validation.md.

Validation rules

• Findings tagged 🟢 verified, 🟡 medium, 🔴 assumed — never collapse into prose.
• Tech debt: every P0/P1 has owner + target date; remediation fits capacity.
• ADRs: 3-year TCO per option, "do nothing" alternative documented, reversibility addressed, repo-committed.
• Build vs buy: must-have/nice-to-have separated; ≥ 2 vendor candidates; ADR committed.
• Health review: all four DORA metrics together; team-satisfaction signal current within 90 days.
• All output passes the Internal Quality Loop (self-verify → peer-verify → critic pre-screen).

Output expectations

Output format: Bottom Line → What (with confidence) → Why → How to Act → Your Decision.

Worked examples for each: references/examples.md.

Metrics dashboard (compact)

Full dashboard with cadences and DORA bands: references/engineering-org-and-process.md.

Context integration

• Always read company-context.md before responding when it exists.
• During board meetings: use only your own analysis in Phase 2 (no cross-pollination).
• Invocation: request input from other roles with [INVOKE:role|question].

Reference map