Full-Stack AI Platform Engineer: Build Safe Agents

Hidden Jobs; United States; Remote; USD 111,000 - 143,000; Full time

Job summary

Thumbtack is hiring to build internal product capabilities that enable AI-enabled agents for essential operational work, focusing on configurable, reviewable, observable, and correctable agent activity so automation can move from experimentation to dependable business workflows.

You will work across the frontend, backend, data, and production lifecycle, partnering directly with teams who use the systems and owning features from design through post-release improvements.

Qualifications

• 3+ years of professional software engineering experience.
• Strong frontend development skills with React and TypeScript, including complex state management and data-dense interfaces.
• Solid backend engineering ability and the capacity to learn unfamiliar languages (Go, PHP, Python).
• Working knowledge of relational and non-relational databases (PostgreSQL, DynamoDB).
• Strong product judgment: identify valuable workflow improvements and decide what not to build.
• Comfort working through ambiguity and delivering in an evolving environment.

Responsibilities

• Design and implement interfaces for configuring agents, reviewing their work, approving actions, correcting mistakes, and understanding outcomes.
• Create observability and evaluation tools to show agent success, failure, or changes over time.
• Develop features end to end across React, TypeScript, service code, and data models.
• Translate ambiguous operational needs from internal business teams into useful software.
• Own features through design, implementation, testing, launch, monitoring, and post-release improvement.
• Design for uncertainty, partial failure, and human intervention rather than assuming every automated result is correct.

Skills

React

TypeScript

Go

PHP

Python

PostgreSQL

DynamoDB

Ambiguity tolerance

Product judgment

Tools

Go

Job description

Thumbtack is hiring to build internal product capabilities that enable AI-enabled agents for essential operational work, focusing on configurable, reviewable, observable, and correctable agent activity so automation can move from experimentation to dependable business workflows.

You will work across the frontend, backend, data, and production lifecycle, partnering directly with teams who use the systems and owning features from design through post-release improvements.