Proxis is the AI automation platform built for enterprises.

Proxis is creating AI email agents that empower brands and enterprise teams in the creator economy to close deals automatically. Think: your inbox, on autopilot.

Proxis initially fine tuned and distilled LLMs for enterprise use—and quickly realized that companies need more than just smarter models. They need full-stack automation. So Proxis built AI agents that handle the tedious, time-consuming work—everything from autonomous sales closers to customer service agents.

The mission is bold: to build the world’s most capable team of AI agents for every enterprise. Proxis imagines a future where a single founder can run a billion-dollar company—backed by a custom-built army of agents. Proxis is rethinking how agents are built, trained, and deployed from the ground up.

As employee #1 you will be working directly with the CEO to build out Proxis’s mission of enabling agentic automation for non-technical enterprises. This will be hard work - it takes a lot to achieve the lofty goals we have set for ourselves. We are looking for people who are hungry, and who want to come build a meaningful company. We want people who are looking to leave their impact.

We are a fully remote company and are accepting applicants from both the US and abroad. We are after the best talent no matter where you are in the world.

We are looking for ML engineers who have hands-on experience with shipping models to production, working with RAG pipelines, and fine-tuning models. The stack you will be working with is Python and TypeScript, with a web platform for orchestrating multi-agent workflows and a job system to manage their execution.

The day to day will be evaluating model performance, identifying and implementing model improvements and orchestrating the infrastructure around our web app. Familiarity with job systems, AWS, Postgres, and Supabase is ideal.

If you are hungry and you are looking to empower others through AI agents then reach out. We want you on board.

The Problem

Training AI agents is hard. Multi-agent orchestration is hard. Enterprises want AI but don't know how to implement it.

The Proxis Solution

That's where Proxis comes in. We are reimagining the way non-technical teams train and use AI agents, starting with email. Our agent learns how you edit drafts, pulls in context from your full email history and company context and allows non-technical teams to refine their own agent.