Full-Stack AI Engineer

About SelfInspection

SelfInspection is an AI-powered vehicle inspection platform serving rental companies, insurers, and fleet operators. We're on a mission to transform how damage is captured, assessed, and acted on — replacing manual processes with intelligent, real-time computer-vision pipelines and seamless API integrations. This is a pivotal moment: we're moving from a product that works to a product that thinks.

The Role

We're looking for a Full-Stack AI Engineer who is equally comfortable building a polished React UI, designing a NestJS microservice, and wiring together an LLM-powered automation pipeline. You'll be a key driver in our transition from traditional software development to an AI-forward engineering culture — introducing agents, RAG systems, and intelligent tooling that multiplies the output of the entire team.

This isn't a role for someone who wants to sit in one lane. You'll move between the frontend, the API layer, ML pipelines, and AI agent workflows depending on what the team needs most.

Core Responsibilities

Full-Stack Product Development

• Build and maintain React/TypeScript frontend features across the inspection and reporting surfaces
• Design and develop NestJS REST APIs and event-driven backend services
• Integrate with external partners via well-documented API contracts and webhooks
• Own features end-to-end: design, implementation, testing, and deployment

AI Automation & Agents

• Design and ship AI agents for SDLC automation: unit test generation, PRD-to-stories, PR review, release notes, and bug-to-fix-to-deploy pipelines
• Build and maintain RAG systems over internal documentation, codebases, and inspection data
• Develop MCP servers to give AI agents structured access to company data and tools
• Implement automated test suites using Playwright and AI-assisted test generation
• Champion the adoption of AI tooling (Cursor, Claude, Codex, GitHub Copilot) across the engineering team

ML Pipeline & Computer Vision

• Contribute to end-to-end ML pipelines: data ingestion, preprocessing, model training, evaluation, and serving
• Integrate computer-vision model outputs into product APIs and reporting workflows
• Write Python scripts and services for data processing, annotation tooling, and model evaluation
• Collaborate with ML teammates on dataset management, labeling pipelines, and model versioning

Platform & Process Improvement

• Drive the shift from manual workflows to AI-augmented ones — identify automation opportunities and build them
• Write and maintain technical specs using OpenAPI/Swagger, BMAD methodology, and spec-kit tooling
• Instrument code for observability; participate in incident response and postmortems
• Mentor teammates on AI tooling best practices and document internal frameworks

Required Skills & Experience

Languages & Frameworks: React, TypeScript, NestJS, Node.js, Python, JavaScript (ES2022+)

AI / ML Tooling: Claude / Anthropic API, OpenAI / Codex, Cursor, LangChain, RAG Systems, MCP

Process & Spec: OpenAPI / Swagger, BMAD, Spec Kit, Playwright, Jest / Vitest

Infrastructure & Ops: AWS (S3, Lambda), Docker, PostgreSQL, REST / Webhooks, CI/CD

Minimum Qualifications

• 5+ years of professional software engineering experience across frontend and backend
• Shipped production React applications and NestJS (or equivalent Node.js) services
• Hands-on experience building with at least one major LLM API (Claude, OpenAI, Gemini)
• Experience writing and running automated test suites (unit, integration, E2E)
• Comfortable reading and authoring OpenAPI specs
• Strong written communication — can write a clear technical spec or architecture decision record

Nice to Have

• Experience with BMAD methodology or spec-kit-style development frameworks
• Familiarity with computer vision workflows, annotation pipelines, or MLOps tooling
• Prior work building MCP servers or structured tool-use integrations for AI agents
• Contributions to AI agent orchestration frameworks (LangGraph, CrewAI, AutoGen, etc.)
• Experience in insurtech, automotive, or fleet/rental tech verticals
• Background shipping AI-powered features to end users, not just internal tooling

How We Work

• Small, focused engineering team — high ownership, low bureaucracy
• AI-first development culture: we use Claude, Cursor, and AI agents in daily workflows
• Async-friendly with weekly syncs; documentation over tribal knowledge
• Jira for project tracking; Slack for team communication
• We ship to staging often and to production deliberately

Compensation & Benefits

• Competitive salary commensurate with experience
• Equity participation
• Flexible remote-first work environment
• Learning & development budget — courses, conferences, and AI tool subscriptions covered