Software Engineer, AI Tooling

New

SupabaseAI developer tools

We believe you can do your best work from anywhere.Full-TimeSenior

Salary not disclosed

Apply NowOpens the employer's application page

Job Details

Experience: 5+ years of experience in professional software engineering experience.
Required Skills: PostgreSQLJavascriptTypeScript

Have 5+ years of professional software engineering experience.
Write strong TypeScript and know the JavaScript ecosystem, including npm, ES modules, bundlers and compatibility.
Be deeply familiar with current AI tools, agent patterns and practical trade-offs.
Have built developer tools, product interfaces or platform primitives used by other engineers.
Have designed or used evaluation frameworks to measure AI quality, reliability and task success, and understand baselines, variance and cost.
Have strong judgment about where AI improves workflows and how to avoid fragile or gimmicky experiences.
Write tests and validate reliability in real-world conditions.
Know Postgres and relational database practice.
Communicate clearly in writing and work across product, engineering and docs.
Be comfortable navigating ambiguity and moving quickly.
Nice to have: experience with vector databases and embeddings, ideally pgvector.
Nice to have: experience with Braintrust or a similar evaluation and tracing platform, or setting up CI quality gates.

Build and improve the MCP server, agent skills, agents and dashboard assistant.
Create tools, skills and evaluations that make new Supabase products usable by agents.
Prototype and evaluate tooling for emerging AI technologies, and help decide what should go to production.
Build evaluations, instrumentation and feedback loops; measure tools against a no-tool baseline and help define quality KPIs and release gates.
Use customer and agent behavior to improve tooling, interfaces and documentation.
Design how documentation and product knowledge are exposed to agents, and measure whether it works.
Build patterns and tools for customers adding AI to their apps, including MCP on Edge Functions, vector embeddings and agents running on Edge Functions.
Enable other teams to contribute to AI Tooling through contribution paths, templates, documentation and review practices.
Centralize shared AI logic and ship tools with tests, benchmarking and stress testing.
Work with AI Tooling Infrastructure engineers on reliability, auth and hosting, and collaborate with Docs, CLI, Frontend, SDK and other teams.

View Full Description & ApplyYou'll be redirected to the employer's site