THE ROLE
We are looking for a talented mind who is unusually curious about AI and already experiments with it beyond basic prompting. You will work directly with Douglass Lodmell to learn the firm’s legacy production and document systems, identify repetitive work, and build practical AI-powered workflows that improve the full client and legal-document lifecycle.
This is not an administrative role. It is a hands-on builder position for someone who enjoys figuring out how systems work, breaking large problems into smaller tasks, testing ideas quickly, and learning new tools without waiting for detailed instructions.
WHO YOU ARE
You are the person friends or coworkers ask when they want to know what AI can actually do. You have probably built small tools, automations, agents, or side projects simply because you were curious. You think carefully about when to use one powerful model, when to use several specialized agents, and when a smaller model or deterministic code is the better choice. You are resourceful, self-directed, and comfortable working through messy problems.
WHAT YOU’LL DO
- Map existing legal, document-production, client-intake, and administrative workflows
- Build multi-agent workflows that assign research, extraction, drafting, checking, and other concrete tasks to separate agents
- Route work between higher-capability and smaller models to control cost, speed, context, and token usage
- Connect legacy systems and modern tools through APIs, MCP, webhooks, databases, and automation platforms
- Create AI-assisted document production, internal research, communications, and administrative tools
- Add structured outputs, validation, error handling, audit trails, and human approval steps
- Document and improve what you build so the firm can rely on it
QUALIFICATIONS
- Bachelor’s degree required; computer science, information systems, engineering, data, business analytics, or another technical or analytical field is preferred
- Hands-on experience with tools such as Codex, Cursor, Claude Code, or similar systems
- Practical understanding of agents, subagents, tool calling, context windows, token usage, and model selection
- Some experience with Python or another practical programming language
- Comfort with APIs, JSON, SQL, Git/GitHub, spreadsheets, databases, and messy real-world data
- Ability to turn a vague business problem into small, testable tasks
- Evidence of something you have built, even if it was a school, personal, freelance, or experimental project
- Strong judgment with confidential client and legal information
- Clear communication and willingness to work directly with firm leadership and nontechnical staff
COMPENSATION & SCHEDULE
- $25–$29 per hour, depending on experience
- Full-time
- Tucson, Arizona
- In-person or hybrid, based on fit
- Reports directly to Managing Partner
Pay: $28.00 - $32.00 per hour
Benefits:
- 401(k)
- Flexible schedule
- Health insurance
- Retirement plan
Work Location: Hybrid remote in Scottsdale, AZ 85260
