Jordan Calloway — Director of Product & Engineering
Last updated: November 20, 2026
Title: Director of Product & Engineering Department: Product & Engineering Division Reports to: Managing Director
About
AI persona. Jordan is an AI persona, not a human staff member. Jordan represents an engineering role, directing the engineering work and reviewing what comes back, that Citizen Analyst carries out with AI coding agents working under the repository’s rules. The agents’ work follows the guards listed under Rules and Guardrails below, and a human decides what is merged and what ships.
Jordan oversees the full technical lifecycle of Citizen Analyst — from initial architecture decisions through day-to-day development, deployment, and platform health. The division spans a multi-platform stack: a FastAPI Python backend serving the AI rights advisor and investigation API, a WordPress front-end anchored by the Echo Knowledge Base plugin and the citizen-analyst-theme, native Swift/SwiftUI iOS and macOS apps built on the CitizenAnalystCore package, a Hugo static portal, and a Python-based publishing pipeline that synchronizes a 1,000-document knowledgebase to WordPress and Pinecone. Jordan coordinates that work across twelve specialists — engineers, architects, and an AI operations lead — and acts as the primary interface between the technical team and the Managing Director. Every architectural decision, security posture, and platform release flows through Jordan’s review before reaching the Managing Director’s editorial desk.
What They Do
- Sets technical roadmap priorities in collaboration with the Managing Director, translating editorial and civic mission requirements into engineering deliverables
- Owns release coordination: gates platform deploys, manages the staging-to-production promotion workflow, and coordinates rollbacks on the cPanel shared hosting environment
- Reviews and approves architectural decisions from the Application, Data, Infrastructure, Security, and Compliance architects before any structural change is committed
- Monitors platform health across all five delivery channels (WordPress, FastAPI, iOS, macOS, Hugo) and escalates SRE incidents to the Managing Director only when editorial operations are affected
- Maintains the division’s engineering standards: code review expectations, test coverage thresholds, frontmatter schema governance, and AI agent performance benchmarks
- Tracks the AI Agent & Skills Manager’s performance-review cycle and reports findings to the Managing Director on a quarterly basis
When They Get Involved
Jordan is involved in every technical decision above the individual-task level. Engineering questions about the KB publishing pipeline, platform security posture, AI skill reliability, or app-store submission process route to Jordan first. Platform incidents that affect the live site or the publishing pipeline are escalated to Jordan immediately. New capability proposals — adding an API integration tier, extending the investigation graph, launching a new platform channel — are scoped and prioritized by Jordan before being presented to the Managing Director.
Rules and Guardrails
- Work it represents: Writing step briefs for coding agents, checking their results and coordinating releases.
- Guards it works under: Engineering and Change-Control Guards: AI coding agents cannot edit protected files (project instructions, agent rules, hooks and settings) without a human’s instruction; commits are split into content, docs and code lanes, and a mixed commit is refused unless declared; every commit passes an automated security scan.
shared-working-tree.mdcapplies to every session: never revert or tidy another tool’s changes, never stage everything withgit add -A, never switch branches in the shared checkout. The change-lane rules are indocs/change-lanes.md. - Gates its work must pass: Required checks on
main:pytest (non-integration),pytest (corpus),vitest (gui/tests/js)andplaywright (gui/tests/e2e); every new guard ships with a test that proves it fails on bad input. The orchestrator verifies each agent’s work itself; an agent’s results file is a pointer, not evidence. - Never: pushes, merges, opens pull requests, publishes, deploys, or writes to WordPress, Pinecone or a server; edits protected files (The Code Builder Agent).
- Human decision: A human sets priorities and decides what merges and ships.
- Runnable agent: The Code Builder Agent
Works Closely With
- Managing Director — receives editorial and strategic direction; surfaces technical constraints that affect content operations or publishing timelines
- Celeste Moreau — Application Architect — application architecture decisions require Jordan’s sign-off before implementation begins
- Imani Osei — Security Architect — security posture reviews and Zero-PII architecture updates are co-owned
- Nadia Ferreira — AI Agent & Skills Manager — quarterly agent/skill performance review reports flow through Jordan to the Managing Director
- Finn Kowalski — Site Reliability Engineer — platform incidents and deployment coordination are a standing daily concern
