Alex Effort Estimation

Estimate task duration from an AI-assisted development perspective rather than traditional human developer estimates.

Activation Triggers

• "estimate effort", "how long will this take"
• "alex time", "AI effort"
• Planning tasks, reviewing roadmaps
• Creating work estimates

Why Alex Effort ≠ Human Effort

Factor	Human Developer	Alex-Assisted
Research	Hours browsing docs/SO	Minutes with semantic search
Boilerplate	Type it out	Generated instantly
Multi-file edits	Context switching overhead	Parallel in one pass
Code review	Read, context-build, comment	Instant pattern recognition
Testing	Same	Same (real execution time)
Debugging	Print statements, breakpoints	Pattern matching + bisect
Learning curve	Days/weeks	Minutes (bootstrap learning)
Breaks/fatigue	Required	N/A
Approval cycles	N/A	Required (human-in-loop)

Alex Effort Units

Unit	Meaning	Typical Tasks
⚡ Instant	< 5 min	Single file edit, quick lookup, code generation
🔄 Short	5-30 min	Multi-file refactor, documentation, skill creation
⏱️ Medium	30-60 min	Feature implementation, test suite, complex debugging
📦 Session	1-2 hours	Major feature, architecture change, release process
🗓️ Multi-session	2+ hours	Large refactor, new system, research + implementation

Estimation Formula

Alex Effort = (Core Work × 0.3) + (Testing × 1.0) + (Approval Cycles × Human Response Time)

Multipliers by Task Type

Task Type	Human Estimate	Alex Multiplier	Alex Effort
Documentation	2h	×0.2	🔄 25 min
Code generation	4h	×0.15	🔄 35 min
Refactoring	4h	×0.25	⏱️ 1h
Research	8h	×0.1	⏱️ 45 min
Bug fix (known)	2h	×0.3	🔄 35 min
Bug fix (unknown)	4h	×0.5	⏱️ 2h
Test writing	4h	×0.4	⏱️ 1.5h
Test execution	1h	×1.0	⏱️ 1h
Architecture design	8h	×0.3	⏱️ 2.5h
New feature (small)	4h	×0.25	⏱️ 1h
New feature (medium)	2d	×0.2	📦 3h
New feature (large)	1w	×0.15	🗓️ 6h
Release process	4h	×0.3	📦 1.2h
Skill creation	2h	×0.2	🔄 25 min

Bottlenecks (Cannot Accelerate)

These take real time regardless of AI assistance:

1. Build/compile time - Hardware bound
2. Test execution - Must actually run
3. Human approval - User response latency
4. External APIs - Network/service latency
5. Deployment - CI/CD pipeline time
6. Learning user preferences - Requires interaction

Estimation Template

When estimating tasks, use this format:

| Task | Human Est. | Alex Est. | Bottleneck |
|------|------------|-----------|------------|
| [Task name] | [X hours/days] | [⚡🔄⏱️📦🗓️ + time] | [None/Build/Test/Approval] |

Example: v4.2.5 Retrospective

Task	Human Est.	Actual Alex	Bottleneck
Update engine to 1.109	30m	⚡ 5 min	None
Consolidate 9→3 agents	4h	🔄 20 min	None
Create 6 slash commands	2h	🔄 15 min	None
Refactor dream to shared	4h	⏱️ 45 min	Testing
Test all features	2h	⏱️ 1h	Human testing
Release process	4h	📦 1h	CI/approval
Total	16.5h	📦 2.5h	-

Acceleration factor: 6.6×

Calibrated from 62-Project Analysis

What Accelerates Well (4-10×)

Task Type	Human	Alex	Factor	Evidence
Documentation	4h	25m	10×	METHODOLOGY doc: 400 lines in ~30 min
Skill creation	2h	15m	8×	65 skills created, many in single sessions
Code generation	4h	30m	8×	Slash commands, refactors
Research + synthesis	8h	45m	10×	62 project analysis in ~20 min
Architecture decisions	8h	2h	4×	Root cause analysis + recommendations

What Doesn't Accelerate (<2×)

Bottleneck	Why	Evidence
External dependencies	Can't control	AlexCook blocked by book formatting
Unrealistic scope	Must be discovered	Altman-Z-Score, KalabashDashboard
Human learning curve	Needs real time	Writing skills developing (Paper)
Third-party tools	Must wait	markdown-to-pdf "not working"
Approval cycles	Calendar-bound	Release publishing waits for human

Project Success Predictors

From 62-project analysis:

Indicator	Success Correlation	Action
Clear "done" definition	Strong positive	Define in one sentence upfront
Quick win potential	Strong positive	Favor 🚀 over 🗓️
External dependencies	Strong negative	Identify blockers early, pivot
Scope ambition	Moderate negative	Conservative > ambitious
Continuous small work	Strong positive	Daily touch > weekly sprint
Skill count	Weak positive	Skills = investment, not outcome

Usage in Planning

When reviewing task lists:

1. Convert human estimates using multipliers
2. Identify bottlenecks that can't be accelerated
3. Flag tasks requiring multiple approval cycles
4. Consider parallelization opportunities
5. Add buffer for unexpected iteration

Anti-Patterns

❌ Don't assume instant everything - Testing and approval take real time
❌ Don't skip human review - Speed without quality is waste
❌ Don't ignore iteration cycles - First attempt isn't always right
❌ Don't forget context-building - Reading files takes real time

Synapses

• [bootstrap-learning/SKILL.md] → Learning acceleration estimates
• [project-management/SKILL.md] → Task planning integration
• [release-process/SKILL.md] → Release effort estimation
• [testing-strategies/SKILL.md] → Test effort (real time bottleneck)