Python Engineer

UMATR; San Francisco (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

UMATR is partnering with a company hiring an AI Engineer to build backend infrastructure that unifies access to AI model providers and simplifies developer integration with LLMs.

You will primarily work in Python with FastAPI, async frameworks, and multiple provider SDKs, creating a unified API layer that normalizes calls and enhances developer experience for AI services.

Qualifications

• 1–2+ years of experience building and shipping backend or full-stack software
• Production APIs with FastAPI or similar async frameworks
• Hands-on experience with LLM APIs (OpenAI/Anthropic)

Responsibilities

• Develop and maintain connectors to multiple LLM and AI model APIs
• Design normalization layers to unify provider API formats
• Build abstractions across providers for sessions, batching, and context handling
• Architect systems exposing advanced provider features with a consistent developer API
• Work with LLMs, model-context protocols, and AI agent systems
• Implement authentication flows and integrations across providers and MCP environments
• Contribute to Python-based APIs and backend services
• Focus on developer experience, reliability, and intuitive APIs
• Engage with users to gather feedback and translate use cases into improvements

Skills

Backend development

FastAPI

LLM APIs

API design

Redis

PostgreSQL

OpenAI API

Anthropic API

Education

Top 25 university or founder experience

Tools

OpenAI API

Anthropic API

Job description

We’re partnered with a company looking to hiring an AI Engineer to build backend infrastructure that unifies access to multiple AI model providers and simplifies how developers integrate with LLMs and AI agents.

You’ll work primarily in Python with modern async web frameworks and multiple LLM provider SDKs, building the layer that normalizes calls across different model APIs and shapes how developers interact with AI services.

This is a hands‑on engineering role where you’ll collaborate closely with senior technical leadership, influence core product direction, and see your work adopted by a broad developer audience.

What you’ll do

• Develop and maintain connectors to multiple LLM and AI model APIs
• Design normalization layers that convert between different provider API formats
• Build abstractions that work across providers for common patterns like sessions, batching, and context handling
• Architect systems that expose advanced provider features while keeping a consistent developer API
• Work with LLMs, model-context protocols, and AI agent systems, prioritizing clean APIs and developer ergonomics
• Implement authentication flows and integrations across different AI providers and MCP environments
• Contribute to the design and development of Python-based APIs and backend services
• Think carefully about developer experience and how to make APIs simple, reliable, and intuitive
• Work directly with users to understand problems, gather feedback, and turn real‑world use cases into product improvements

What we’re looking for

• 1-2+ years of experience building and shipping backend or full‑stack software
• Experience building production APIs with FastAPI or similar async frameworks
• Hands‑on experience working with LLM APIs, particularly OpenAI or Anthropic
• Solid understanding of API design, request/response structures, and provider‑specific API differences
• Comfortable working with Redis and PostgreSQL
• Genuine interest in LLMs, AI agents, model‑context protocols, and developer infrastructure
• Comfortable speaking directly with users and turning their feedback into technical solutions
• Strong background from a top 25 university or demonstrable track record as a founder/founding engineer at an early‑stage startup