AI Agent Research Engineer

Join Autonomy AI to shape the future of autonomous AI agents built to assist developers. As an AI Researcher / Agent Engineer, you’ll work at the cutting edge of language models, reasoning systems, and intelligent agents. Help design novel agent architectures, iterate on prompting strategies, and build intelligent systems that reason about code, tools, and context to enhance software development.

Why join us?

Build foundational technology in one of the fastest-moving fields in AI.
Tackle real-world challenges in reasoning, planning, and autonomous tool use.
Work alongside an elite team of engineers, researchers, and product thinkers.
Enjoy a high-ownership, low-ego environment where your ideas make a direct impact.
Join a well-funded, high-growth startup with massive technical upside.

Our Tech Stack

Python, OpenAI, Claude, local LLMs, GRPC, Redis, Postgres, AWS, and custom agent runtime environments.

Requirements:

Responsibilities:

As an AI Agent Research Engineer, you will design, prototype, and deploy autonomous agents that interact with real-world software systems. You’ll conduct research on multi-step reasoning, RAG, tool use, and long-term memory strategies, working to optimize agent behavior through advanced prompting techniques, context modeling, and architecture-level innovation. A key part of your role will be leveraging and extending our existing infrastructure to ensure agents remain generalizable across diverse codebase structures and development environments. You’ll collaborate closely with product and engineering teams to bring experimental ideas into production, and analyze agent behavior at scale to identify failure modes, edge cases, and performance bottlenecks.

Preferred Qualifications:

2+ years of experience in applied AI, ML research, or systems involving language models.
Solid experience as a software engineer, with preference for full-stack development background to ensure strong domain understanding.
Proven experience in data science and algorithmic development.
Deep understanding of transformer-based LLMs, vector retrieval, and prompt engineering.
Familiarity with building agents is a plus, but not required—we will support your training in this emerging space.
Experience with Python and working with LLM APIs or fine-tuning pipelines.
Familiarity with tools like Docker, cloud environments (AWS/GCP), and scalable infrastructure.
Advantage: Experience in developer tooling, compilers, or software engineering workflows.
Advantage: Research or publications in AI planning, LLM agents, or program synthesis.