Case Study
SEO Wires
Full-Stack Developer at BlackWires | Feb 2026 - Present
My Role
SEO Wires is a staged content-research system that starts with a website and produces competitor analysis, keyword opportunities, topic groups, and article drafts. At BlackWires, I built the queue-driven worker pipeline behind that workflow.
What I Built
Site Crawl
Built the worker that crawls a submitted website and extracts the content needed to understand the business and its existing coverage.
Competitor Discovery
Built the stage that identifies the 15 closest search competitors and passes their relevant pages into analysis.
Keyword Gap Analysis
Implemented the processing stage that compares site and competitor coverage, then groups opportunities into usable topics.
Research to Draft
Built the writing stage that combines search research, outline context, site voice, and internal-link opportunities into a draft.
Technical Decisions
One Queue per Stage
Crawling, competitor discovery, gap analysis, and writing run as separate workers with explicit handoffs.
Retry and Backoff
BullMQ retries transient failures with backoff so a stalled crawl or provider request does not discard the full pipeline.
Stage Isolation
Each worker can fail, retry, and scale independently without forcing the entire analysis to restart.
Asynchronous Requests
Long audits leave the web request cycle immediately, allowing the application to remain responsive while workers continue processing.
Container Deployment
Workers and application services run in Docker on AWS so processing capacity can be managed independently.
Outcome
The pipeline passes work from crawl to draft without blocking application requests or requiring manual data transfer between stages. Independent workers and retry policies make long-running analyses easier to recover and operate in production.