메뉴
Infrastructure and development cost estimation for technical projects. Use when planning budgets, evaluating build vs buy decisions, or projecting TCO for architecture choices.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | cost-estimator |
| description | Infrastructure and development cost estimation for technical projects. Use when planning budgets, evaluating build vs buy decisions, or projecting TCO for architecture choices. |
Provides frameworks for estimating infrastructure costs, development effort, and total cost of ownership (TCO) for technical projects.
TCO = Infrastructure + Development + Operations + Opportunity Cost
┌─────────────────────────────────────────────────────────────────┐
│ TOTAL COST OF OWNERSHIP │
├─────────────────────────────────────────────────────────────────┤
│ │
│ Infrastructure Development Operations Opportunity │
│ ──────────────── ──────────── ────────── ──────────── │
│ • Compute • Engineering • Support • What else │
│ • Storage • QA • Monitoring • could team │
│ • Network • DevOps • On-call • be building? │
│ • Third-party • Management • Training │
│ APIs/SaaS • Contractors • Incidents │
│ │
└─────────────────────────────────────────────────────────────────┘
| Instance | vCPU | RAM | Monthly Cost | Best For |
|---|---|---|---|---|
| t3.micro | 2 | 1GB | $8 | Dev/test |
| t3.medium | 2 | 4GB | $30 | Small apps |
| t3.large | 2 | 8GB | $60 | Light production |
| m6i.large | 2 | 8GB | $70 | General production |
| m6i.xlarge | 4 | 16GB | $140 | Medium workloads |
| m6i.2xlarge | 8 | 32GB | $280 | Heavy workloads |
| c6i.2xlarge | 8 | 16GB | $250 | CPU-intensive |
| r6i.2xlarge | 8 | 64GB | $370 | Memory-intensive |
| Instance | GPU | VRAM | Monthly Cost | Best For |
|---|---|---|---|---|
| g4dn.xlarge | T4 | 16GB | $380 | Inference |
| g5.xlarge | A10G | 24GB | $730 | ML training/inference |
| p4d.24xlarge | 8x A100 | 320GB | $23,000 | Large model training |
| Plan | Savings | Commitment |
|---|---|---|
| On-Demand | 0% | None |
| Reserved (1yr) | 30-40% | 1 year |
| Reserved (3yr) | 50-60% | 3 years |
| Spot Instances | 60-90% | Can be interrupted |
| Instance | vCPU | RAM | Monthly Cost | Connections |
|---|---|---|---|---|
| db.t3.micro | 2 | 1GB | $15 | 50 |
| db.t3.medium | 2 | 4GB | $50 | 100 |
| db.m6g.large | 2 | 8GB | $120 | 200 |
| db.m6g.xlarge | 4 | 16GB | $240 | 400 |
| db.r6g.xlarge | 4 | 32GB | $350 | 500 |
| db.r6g.2xlarge | 8 | 64GB | $700 | 1000 |
Add for storage: $0.115/GB/month (gp3) Add for IOPS: $0.02/IOPS/month (over 3000 baseline)
| Node Type | RAM | Monthly Cost |
|---|---|---|
| cache.t3.micro | 0.5GB | $12 |
| cache.t3.medium | 3GB | $50 |
| cache.m6g.large | 6.4GB | $100 |
| cache.r6g.large | 13GB | $175 |
| Service | Cost | Use Case |
|---|---|---|
| S3 Standard | $0.023/GB | Frequently accessed |
| S3 Infrequent | $0.0125/GB | Backups, archives |
| S3 Glacier | $0.004/GB | Long-term archive |
| EBS gp3 | $0.08/GB | Block storage |
| EBS io2 | $0.125/GB + IOPS | High performance |
| Traffic Type | Cost |
|---|---|
| Data IN | Free |
| Data OUT (first 10TB) | $0.09/GB |
| Data OUT (next 40TB) | $0.085/GB |
| Inter-AZ transfer | $0.01/GB each way |
| Inter-region transfer | $0.02/GB |
| CloudFront to internet | $0.085/GB |
Development Cost = (Hours × Hourly Rate) × Complexity Factor × Risk Buffer
Hourly Rate (Fully Loaded):
- Junior Engineer: $75-100/hr
- Mid-level Engineer: $100-150/hr
- Senior Engineer: $150-200/hr
- Staff/Principal: $200-300/hr
Complexity Factors:
- Greenfield, known tech: 1.0x
- Existing codebase, known tech: 1.2x
- New technology for team: 1.5x
- Complex integrations: 1.3x
- Regulatory/compliance: 1.4x
Risk Buffer:
- Well-defined requirements: 1.2x
- Ambiguous requirements: 1.5x
- Experimental/R&D: 2.0x
| Size | Story Points | Hours | Cost (Mid-level) |
|---|---|---|---|
| XS | 1 | 2-4 | $200-400 |
| S | 2 | 4-8 | $400-800 |
| M | 3 | 8-16 | $800-1,600 |
| L | 5 | 16-32 | $1,600-3,200 |
| XL | 8 | 32-64 | $3,200-6,400 |
| XXL | 13+ | 64+ | $6,400+ |
## Monthly Team Cost
Engineering Team:
- 2 Senior Engineers × $15,000 = $30,000
- 3 Mid-level Engineers × $10,000 = $30,000
- 1 Engineering Manager × $18,000 = $18,000
Overhead (benefits, tools, etc.): 30%
Monthly Burn: ($78,000) × 1.3 = $101,400
Annual Team Cost: ~$1.2M
Build vs Buy Decision Matrix:
LOW Differentiation HIGH Differentiation
┌────────────────────┬────────────────────┐
HIGH Volume/ │ │ │
Usage │ Consider │ BUILD │
│ Build │ (competitive │
│ (cost savings) │ advantage) │
├────────────────────┼────────────────────┤
LOW Volume/ │ │ │
Usage │ BUY │ BUY │
│ (no question) │ (then consider │
│ │ build if scales) │
└────────────────────┴────────────────────┘
## Option A: Build Custom Solution
### Initial Development
- Engineering time: X months × $Y/month = $Z
- Infrastructure setup: $A
### Ongoing Costs (Annual)
- Infrastructure: $B
- Maintenance (20% of dev time): $C
- On-call/support: $D
### 3-Year TCO
Year 1: $Z + $A + $B + $C + $D
Year 2: $B + $C + $D
Year 3: $B + $C + $D
Total: $XXX
---
## Option B: Buy SaaS Solution
### Initial Costs
- Implementation/integration: $X
- Training: $Y
### Ongoing Costs (Annual)
- License fees: $Z/year
- Per-user costs: $A × users
- API costs: $B
### 3-Year TCO
Year 1: $X + $Y + $Z + $A + $B
Year 2: $Z + $A + $B
Year 3: $Z + $A + $B
Total: $XXX
| Capability | Build When | Buy When |
|---|---|---|
| Authentication | Unique security requirements | Standard OAuth/OIDC works |
| Payments | Core business differentiator | Standard e-commerce |
| Search | Domain-specific relevance | Generic search needs |
| Analytics | Proprietary insights needed | Standard dashboards work |
| High volume, custom delivery | Standard transactional | |
| ML/AI | Proprietary models needed | Pre-trained models work |
| Scale | Users | Monthly Infra | Notes |
|---|---|---|---|
| Startup | 0-1K | $200-500 | Single server, managed DB |
| Growth | 1K-10K | $500-2,000 | Load balancer, caching |
| Scale | 10K-100K | $2,000-10,000 | Horizontal scaling |
| Enterprise | 100K-1M | $10,000-50,000 | Multi-region, HA |
| Large | 1M+ | $50,000+ | Global, custom CDN |
| Application Type | Cost/User/Month | Notes |
|---|---|---|
| Simple web app | $0.05-0.20 | Static + API |
| Data-intensive | $0.20-0.50 | Analytics, storage |
| Real-time | $0.50-2.00 | WebSockets, streaming |
| ML-powered | $1.00-5.00 | Inference costs |
| Video/media | $2.00-10.00 | Transcoding, CDN |
## Monthly Infrastructure Cost by GMV
$0-100K GMV/month:
- Basic infrastructure: $500
- Payment processing (2.9%): ~$2,000
- Total: ~$2,500
$100K-1M GMV/month:
- Scaled infrastructure: $2,000
- Payment processing: ~$20,000
- Fraud protection: $500
- Total: ~$22,500
$1M-10M GMV/month:
- HA infrastructure: $10,000
- Payment processing: ~$200,000
- Fraud/security: $5,000
- CDN/performance: $3,000
- Total: ~$218,000
Infrastructure:
Development:
Operations:
Third-Party Services:
| Strategy | Savings | Effort |
|---|---|---|
| Reserved instances | 30-60% | Low |
| Right-sizing instances | 20-40% | Medium |
| Spot instances (non-critical) | 60-90% | Medium |
| Storage tiering | 50-80% | Low |
| CDN caching | 30-50% bandwidth | Low |
| Optimization | Impact | Complexity |
|---|---|---|
| Caching (Redis) | 50-80% DB load reduction | Medium |
| Queue-based processing | Smooth traffic spikes | Medium |
| Auto-scaling | Pay for what you use | Medium |
| Serverless (appropriate use) | Variable → zero when idle | High |
| Multi-region read replicas | Reduce cross-region costs | High |
# Project: [Name]
# Duration: [X months]
## Development Costs
| Phase | Duration | Team Size | Cost |
|-------|----------|-----------|------|
| Discovery/Design | 2 weeks | 2 | $X |
| MVP Development | 8 weeks | 4 | $X |
| Testing/QA | 2 weeks | 3 | $X |
| Deployment | 1 week | 2 | $X |
| **Total Development** | | | **$X** |
## Infrastructure Costs (First Year)
| Component | Monthly | Annual |
|-----------|---------|--------|
| Compute | $X | $X |
| Database | $X | $X |
| Storage | $X | $X |
| Network | $X | $X |
| Third-party APIs | $X | $X |
| Monitoring/Tools | $X | $X |
| **Total Infrastructure** | **$X** | **$X** |
## Ongoing Costs (Annual)
| Category | Cost |
|----------|------|
| Infrastructure | $X |
| Maintenance (20% of dev) | $X |
| Support/On-call | $X |
| Tool licenses | $X |
| **Total Annual** | **$X** |
## Summary
| Metric | Value |
|--------|-------|
| Total First Year | $X |
| Annual Run Rate | $X |
| 3-Year TCO | $X |
| Cost per User (at scale) | $X |
## Quick Infrastructure Estimate
Inputs:
- Expected users: [X]
- Requests per user/day: [Y]
- Data storage per user: [Z GB]
- Growth rate: [W%/month]
Calculations:
- Daily requests: X × Y
- Monthly requests: Daily × 30
- Required compute: (Monthly requests / 100K) × $50
- Storage: X × Z × $0.10
- Database: (X / 10K) × $200
- Estimated monthly: Compute + Storage + Database × 1.3
12-month projection with growth:
Sum of (Monthly × (1 + W%)^month) for months 1-12
Detect common technical and organizational anti-patterns in proposals, architectures, and plans. Use when strategic-cto-mentor needs to identify red flags before they become problems.
Recommend architecture patterns (monolith, microservices, serverless, modular monolith) based on scale, team size, and constraints. Use when cto-architect needs to select the right architectural approach for a new system or migration.
Identify and challenge implicit assumptions in plans, proposals, and technical decisions. Use when strategic-cto-mentor needs to surface hidden assumptions and wishful thinking before they become costly mistakes.
Generate targeted clarifying questions (2-3 max) that challenge vague requirements and extract missing context. Use after request-analyzer identifies clarification needs, before routing to specialist agents. Helps cto-orchestrator avoid delegating unclear requirements.
Transform clarified user requests into structured delegation prompts optimized for specialist agents (cto-architect, strategic-cto-mentor, cv-ml-architect). Use after clarification is complete, before routing to specialist agents. Ensures agents receive complete context for effective work.
Deep expertise in ML/CV model selection, training pipelines, and inference architecture. Use when designing machine learning systems, computer vision pipelines, or AI-powered features.