Running systems, not claims

Everything on this page is a live system on this domain — the same hands that build it are available for community, affiliate-partnership and operations roles. If a hiring manager wants proof of execution, this page is the proof.

Case study 01 · Live · Daily schedule

Job-scout automation engine

A scheduled Cloudflare Workers pipeline that sources community-manager and partnerships leads, deduplicates them, scores them with an AI model against a fit profile, and generates ready-to-send application drafts — delivered every morning by email (Resend) and Discord.

Measured smoke runs: 264 raw listings → 40 scored → 1 qualified lead persisted. Stack: Cloudflare Workers, D1, KV, Resend API, Discord webhooks, cron (08:30 CET).

Case study 02 · Live · This website

AI-search-native web build

This site was designed for retrieval by AI answer engines (Google AI Mode, AI Overviews), not just classic SERPs. That means: self-contained question-shaped content chunks that survive “query fan-out” sub-searches; an entity-consistent schema.org graph; answer-first sections (studies show most AI citations come from the top third of a page); and clean crawlable architecture with zero client-side content rendering. The full research brief — including why llms.txt is shipped but not counted on — is published with sources.

Read the research brief →

Case study 03 · Live · This website

Tools engine

The drawdown simulator and position risk calculator are dependency-free vanilla JavaScript — no framework, no build step, no trackers. The same infrastructure powers the upcoming deals tracker and the weekly digest.

Stack

Cloudflare (Workers, Pages, D1, KV, Email Routing) · Resend · Discord webhooks · vanilla HTML/CSS/JS · scheduled crons. Zero-cost infrastructure at current scale, documented runbooks for every moving part.

Working with me: community & affiliate operations, KOL programs, automation for trading educators, or a full content-engine build — arturo@aegiszero.io.