---
title: Architect
category: product
entity_type: skill
price: $15
canonical: https://forgehouse.ai/skills/architect/
lang: en
hreflang_alt: https://forgehouse.ai/tr/skiller/architect/
last_updated: 2026-06-20
---

# Architect

> System design and planning patterns for architecture decisions, PRDs, specifications, and…

A system-design and technical-planning toolkit that takes a feature from fuzzy requirements to a concrete, defensible architecture. It guides requirements gathering, data and API design, scalability decisions, and production readiness: and forces every significant choice through structured tools like C4 modeling, ADRs, blast-radius analysis, and second-order effect thinking.

## Use cases
- Writing a PRD or feature specification before implementation
- Choosing between monolith and microservices for a new system
- Selecting a database and API paradigm for a project
- Documenting an architecture decision so future engineers understand the why
- Designing a caching and indexing strategy for scale
- Running a production-readiness review before launch

## Benefits
- Decisions that survive scrutiny because the trade-offs are written down
- Less premature complexity by applying Occam's razor to every layer
- Fewer surprise outages thanks to blast-radius and second-order analysis
- Faster onboarding because the reasoning behind the system is documented

## What’s included
- A three-step planning workflow from requirements to implementation plan
- A feature-specification template with data model, API, and UI sections
- An Architecture Decision Record template with alternatives and consequences
- Decision guides for monolith-vs-microservices, database, and API paradigm
- Caching and database-optimization patterns with code
- A production-readiness checklist spanning infra, security, and operations

## Who it’s for
Engineering leads, architects, and senior developers who plan systems and document the decisions behind them.

## How it runs
A vague feature idea leaves this skill as a buildable spec with its trade-offs on record. The planning loop runs through six fixed stations:
1. Requirements gathering answers a fixed question set before anything else: what problem this solves, who the users are, what the success metrics are, what the constraints and non-functional requirements look like.
2. Technical design fills the spec template section by section: data model with schema, API contract table per endpoint, UI component list, integration points, security considerations and performance targets.
3. Every versus-decision (monolith vs microservices, REST vs GraphQL vs tRPC, PostgreSQL vs MongoDB vs Redis) gets resolved against the built-in decision matrices and recorded as an ADR with status, alternatives considered and consequences.
4. Second-order effect analysis is mandatory on each decision: SSR improves SEO but raises server cost and capacity planning; caching cuts latency but creates invalidation complexity and stale-data risk.
5. The implementation plan breaks into granular tasks with dependencies identified, a risk table with mitigations, and a testing strategy split into unit, integration and E2E layers.
6. Before production, the readiness checklist sweeps infrastructure, application, security, performance and operations items; red flags like unclear requirements or a missing rollback procedure stop the ship.

## FAQ
### My feature is small, is a full architecture toolkit too heavy?
Scale the depth to the decision: a small feature might need only the requirements and data-design steps. The structured tools are there for choices that are expensive to reverse, like monolith versus microservices, not every checkbox.

### Does it pick the database and architecture for me?
It forces each significant choice through a structured comparison so the decision is defensible, but the call is yours, made with the trade-offs laid out. It's a thinking framework, not an oracle that hands you 'use Postgres.'

### What do I actually walk away with?
A PRD or specification and the documented reasoning behind your architecture: planning artifacts, not running code. Implementation is a separate step; this makes sure you build the right thing before you build it.

## Price
$15, one-time, no subscription. VAT included.

Related guide: [AI code review and developer workflow](https://forgehouse.ai/guides/ai-code-review/)