[Hiring] Internal Tools & AI Automation Engineer @Somewhere

Role Description

Our work is high-value and detail-heavy: reading returns, pulling figures from client documents, modeling strategies and tracking each client year-round. As our client base and advisory team grow, too much of that still runs on manual effort. This role exists to take that load off our advisors so their time goes to strategy, not data handling.

You will be our first dedicated technical hire, and the role has two halves:

• First, you sit with our founder and advisors, learn how the work actually gets done, and find where time is being lost.
• Second, you build the fix, using AI-assisted development tools such as Claude Code to ship practical internal tools quickly.

The measure of success is simple: tools our team uses every day, and hours given back to client work. The right person cares more about solving the right problem than building the most impressive thing.

Qualifications

• 3+ years of professional software development, with full-stack applications shipped to real users (Python and/or JavaScript/TypeScript)
• Daily, hands-on use of AI coding tools such as Claude Code, Cursor or GitHub Copilot, with examples of what you have built using them
• At least one real project built on an LLM API, such as Anthropic or OpenAI
• Solid experience with APIs and integrations, and with working data out of PDFs, spreadsheets and SQL databases
• Clear spoken and written English, and comfort leading working sessions with non-technical people to understand what they need
• Able to work a U.S. Pacific Time schedule, with a reliable internet connection and backup power
• Sound judgment and discretion with sensitive financial and personal data

Requirements

• Interview the founder and each advisor to map how client work flows from intake through strategy, filing and ongoing monitoring
• Identify where time is lost to manual steps, rework or scattered information, and estimate the hours each fix would save
• Maintain a prioritized build list with the founder, and agree on scope and a definition of done before building
• Build and ship internal tools using AI-assisted development (Claude Code and similar), moving from working prototype to daily use quickly
• Automate client intake and document collection so files arrive organized and complete
• Build tools that extract and structure data from tax documents and returns, reducing manual entry
• Create calculators and models that let advisors compare tax strategies for a client quickly
• Build tracking views that show where every client stands across the year
• Automate recurring client deliverables, such as summaries and strategy reports, from templates
• Design LLM-powered workflows using APIs such as Anthropic's Claude, including prompt design and output testing
• Build in human review wherever accuracy matters, since tax figures must be correct before they reach a client or a return
• Connect the firm's existing systems, such as CRM, document storage, email, calendar and spreadsheets, through APIs or automation platforms
• Handle client financial and personal data with care: least-privilege access, no client data in unapproved tools, and clear records of what each tool touches
• Support the firm's information security practices as new tools are introduced
• Train the team on each new tool and gather feedback in the first weeks of use
• Document how each tool works so it can be maintained and extended
• Fix issues promptly and improve tools as workflows and tax rules change

Benefits

• Direct line to the founder.
• Real ownership.
• Visible impact.
• Room to grow.

Ideal Candidate Profile

• A builder first. You would rather ship a simple tool this week than design a perfect one next quarter.
• Curious about how other people work. You ask good questions and listen before proposing a solution.
• Comfortable owning a tool end to end, from the first conversation through daily use and maintenance.
• Precise. You know a wrong number in a tax workflow is costly, so you test, verify and build in review steps.
• Self-directed and organized. You can manage your own priorities as the only technical person on a small team.
• A clear communicator who can explain technical trade-offs to a founder in plain language.
• Energized by AI tools and always looking for a faster way to get things done.