Douglas Joseph P.Private profile: Upwork shows its earnings only to signed-in members, so we check it every few weeks, not daily. MRR here is a rough estimate (Upwork's six-month earnings ÷ months active). Last read Nov 5, 2026.

I help teams get AI working on their data. Most of my work falls into four areas: • AI agents on your data. Agents that query your database, build recurring reports, flag what changed and draft the next step, with a person approving anything that matters. Underneath is Text-to-SQL, with every query checked before it runs. I led a healthcare claims analytics build that reached about 95% accuracy across 40+ questions reviewed by the client's analysts, cut prompt size from 13K to 7K tokens, and ran on 100K+ claims a day. • AI readiness and data audits. A fixed-scope review of your processes and data: where AI pays off, what has to be fixed first, and what it will cost to run. • Claude and Claude Code for teams. Hands-on training plus custom Claude skills and connectors, so your team's recurring work runs inside your own Claude. • Hardening AI-built apps and automations. A prototype, a vibe-coded app, or an n8n, Zapier or Coda setup that keeps breaking, made secure, tested, monitored and owned by you. Also: document extraction (PDFs, contracts, statements into structured data), pipelines into PostgreSQL, BigQuery, Snowflake or Databricks, and web data collection for lead enrichment and monitoring. How I work: a short call and a written plan, then small testable steps with recorded walkthroughs. Anything customer-facing gets a person approving it, and every AI change is tested against the previous version before release, so it can be rolled back. You own the code, accounts and documentation at handoff. For larger builds I bring in a senior data engineer from my team. Most engagements continue as retainers. Industries: healthcare and clinical research, legal, finance, manufacturing, e-commerce, HR tech, B2B sales. Tools: Python, SQL, TypeScript; Claude, Claude Code, OpenAI and AWS Bedrock; MCP; RAG and vector databases; PostgreSQL, BigQuery, Snowflake, Databricks; AWS, GCP, Azure, Terraform; n8n, Zapier, Coda; Firecrawl and browser-use. Outside work I thru-hike long trails, including the Appalachian and Pacific Crest Trails. Send me the process you want automated and a sample of the data. I'll tell you honestly whether it's worth building, buying a tool for, or leaving alone.