All projects

Service Desk Content Operations Studio

A Python and SQLite operating system for turning IT service desk expertise into a repeatable, measurable publishing pipeline.

PythonAutomationAI
Private — architecture and implementation details available on request11 post pipeline

The problem

Content ideas, drafts, artwork, publishing status, and engagement data were scattered across folders and spreadsheets. That made a leadership content program hard to operate consistently: status went stale, dates could be malformed, and lessons from published work did not reliably feed the next cycle.

The solution

I built a GitHub-backed Content Studio with a Streamlit dashboard and SQLite source of truth. It manages editable content records, platform status, publish dates, notes, completion state, follower snapshots, engagement checkpoints, backups, and synchronized CSV exports. I paired the app with documented writing and design standards plus Python utilities for generating post workspaces, collecting metrics, and analyzing results.

Impact

  • Converted an 11-post season plan into one visible operating pipeline
  • Kept editable SQLite state and versionable CSV snapshots synchronized
  • Published a complete 10-slide pilot across two social platforms using the standardized workflow
  • Added tested validation and backup paths instead of relying on a brittle spreadsheet

How it works

Architecture / Flow
01Each topic starts in a standardized workspace containing the outline, slide copy, captions, prompts, hashtags, assets, and retrospective
02The Streamlit dashboard validates edits and writes content, platform, metric, and follower records to SQLite
03Successful changes refresh CSV snapshots for portable analysis, recovery, and useful Git history
04Publishing checkpoints and retrospectives feed performance data and lessons back into the next content cycle

Lessons learned

  • Creative work still needs operational discipline: clear states, owners, checkpoints, and a definition of done prevent good ideas from stalling.
  • SQLite works well for live local state, while synchronized CSV snapshots keep the data portable, inspectable, and recoverable.
  • Validation belongs at the point of entry; one malformed date can quietly make an otherwise polished dashboard misleading.
Technologies
PythonStreamlitSQLiteCSV synchronizationUnit-tested data layerGit / GitHub