> Markdown version of [/jobs/ext/4257000-software-engineer-iii-full-stack](https://www.wearedevelopers.com/jobs/ext/4257000-software-engineer-iii-full-stack). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Software Engineer III (Full-Stack)

- **Company:** Sweet James
- **Location:** Newport Beach, CA, United States
- **Experience:** Expert
- **Salary:** $165,000.0 - $200,000.0
- **Contract:** Permanent contract
- **Skills:** Microsoft Windows, Application Programming Interfaces (APIs), Artificial Intelligence, Amazon Web Services, Audit Trail, Automation of Tests, Microsoft Azure, Microsoft Outlook, Cloud Computing, Code Review, Continuous Integration, Customer Data Management, Github, Python (Programming Language), PostgreSQL, Uptime, Node.Js, OpenAI, Next.js, Salesforce.Com, Search Technologies, Microsoft SharePoint, SQL Databases, Twilio, TypeScript, Core Voice Platform, ReactJS, Retrieval-Augmented Generation, Large Language Models, Agentic-AI, Pgvector, Nintex, Graphql, Claude, Model Context Protocol, DocuSign, Docker, Container Apps
- **Published:** October 27, 2026
- **Apply:** https://www.indeed.com/viewjob?jk=c2e81fd426e2db18

## About the Role

Required

 * 6+ years building production software, full-stack: React or Next.js on the front end, Node/TypeScript or Python on the back end.
 * Production SQL and data modeling in Postgres.
 * LLM-powered features shipped to real users: prompt and tool design, structured outputs, retrieval (RAG) and evaluation.
 * Cloud deployment and operations: Docker, Azure or AWS, CI/CD, monitoring, and owning uptime.
 * Experience leading projects and mentoring engineers.
 * Clear writing: design docs, runbooks, and plain-language explanations for people who are not engineers.
 * Sound judgment with sensitive data.

Preferred

 * Anthropic Claude and OpenAI APIs at production scale; agent frameworks or the Model Context Protocol (MCP).
 * Supabase and vector search with pgvector.
 * Workflow automation platforms such as n8n.
 * The Salesforce data model and APIs (Litify a plus).

## Description

Build the apps and AI features our firm runs on, and set the bar for how we build them.

Technology is part of how we practice law. Systems we built and run ourselves read and sort more than 100,000 incoming documents and transcribe more than 100,000 calls every month, chase medical records, and put each case's full history in front of the people working it. We have replaced seven figures of vendor software with our own systems, and we keep adding more.

You are a senior full-stack engineer who builds with AI every day. You will design and ship the internal apps our staff work in, from the screen they click to the service and database behind it, and build AI into them where it saves real time. You lead projects end to end, review other engineers' work, and own what you build in production., * Build full-stack apps staff use all day. React and Next.js front ends, Node/TypeScript and Python services, and Postgres behind them: our records console, claims console, case assistant, review tools and dashboards.

 * Put AI to work inside them. Pull facts out of medical records and police reports, search a case's full history, and run agents that do routine work for a person to approve.
 * Connect the firm's systems. Salesforce/Litify, Outlook and SharePoint (Microsoft Graph), Zoom, Twilio and DocuSign.
 * Lead projects end to end. Turn a request from attorneys or staff into a design, a plan and a shipped result, and measure whether people use it and how much time it saves.
 * Raise the bar on quality. Code review, automated tests, and evaluation sets for every AI feature that measure facts captured, facts missed and wrong answers before a change ships.
 * Run what you build. Deploy containerized services on Azure, monitor them, and fix them fast when they break.
 * Grow the team. Mentor Software Engineers I and II, review their designs and code, and help hire.
 * Protect client data. Least-privilege access, audit trails, and safe handling of medical records (HIPAA) and privileged case information., * 90 days: You own one production app end to end, with tests, monitoring and a runbook.
 * 6 months: You have led a project from design to daily use, and other engineers ship through your reviews.
 * 12 months: Several workflows you built are part of staff's daily routine, with hours saved measured for each.

Our Stack

Python, TypeScript/Node, Next.js/React, Postgres and Supabase with pgvector, Anthropic Claude and OpenAI APIs, Azure Container Apps and Docker, n8n, MCP, Salesforce/Litify, Microsoft 365 and the Graph API, Zoom, Twilio, Retell, GitHub.