As an Applied AI Engineer on the Codex Core Agent team at OpenAI, you’ll work at the forefront of AI development, transforming impressive research demos into dependable, real-world tools. This full-time role focuses on improving agent performance for software engineering tasks, bridging the gap between theoretical capabilities and practical application. It’s an exceptional chance for an early-career professional to contribute directly to the core functionality of a groundbreaking AI product and gain invaluable experience in a leading AI research and deployment company.
Job Snapshot
Company: OpenAI
Role: Applied AI Engineer, Codex Core Agent
Location: New York, NY
Work Type: Full-time
Pay Range: $230,000 – $385,000 / year
🔥 DSFOR Insider Take: Why This Role Matters
Joining OpenAI as an Applied AI Engineer offers a unique opportunity to shape the future of artificial intelligence within one of the most impactful organizations in the field. This role isn’t just about theoretical research; it’s about making AI agents like Codex genuinely functional and useful in real-world scenarios, specifically for complex software engineering tasks. You’ll gain hands-on experience in closing the gap between research and production, working on the core systems that enable AI to perform reliably. This involves optimizing for crucial metrics like production performance (tokens, latency, reliability, cost), refining core execution loops, and building shared infrastructure. The skills you develop in AI system design, performance optimization, and cross-functional collaboration with research and product teams are highly transferable and will position you strongly for future leadership roles in AI development. This experience is particularly valuable for early-career professionals seeking to make a tangible impact on cutting-edge technology.
Role Overview & Responsibilities
On the Codex Core Agent team, you will collaborate closely with research, infrastructure, and product teams to refine the kernel of Codex, ensuring its performance as a dependable agent in real-world applications. Your overarching goal is to transform model capabilities into measurable gains in solve rate and economic value for users.
- Improve agent performance on actual software engineering tasks, bridging the gap between research and practical usefulness.
- Work across systems that enable Codex to function as an agent, focusing on production performance envelopes including tokens, latency, reliability, cost, and capacity.
- Develop and refine the core execution loop and interfaces that translate models into useful agent behavior.
- Build shared infrastructure that empowers other teams to leverage Codex effectively.
- Establish feedback loops to convert real-world usage data into improvements for models and agent behavior.
- Ensure agents are not only powerful but also useful, steerable, and reliable in practice.
- Optimize model behavior to achieve measurable gains in solve rate, usefulness, and economic value for users.
Requirements
- Experience making Codex function effectively as an agent in real-world applications.
- Ability to accelerate AI research and implement improvements in a production environment.
- Proficiency in optimizing production performance across various metrics (tokens, latency, reliability, cost, capacity).
- Skills in developing core execution loops and interfaces to drive useful agent behavior from models.
- A background in building shared infrastructure that supports other teams leveraging complex AI systems.
- A strong understanding of feedback loops for improving AI models and agent behavior based on real-world usage.
📅 What to Expect
You can expect a highly collaborative and fast-paced environment at OpenAI, where you’ll be integrated into a team focused on the practical application of cutting-edge AI. This role involves working closely with diverse teams, including dedicated researchers, infrastructure specialists, and product managers, emphasizing a holistic approach to AI development from conception to deployment. The work is deeply focused on engineering robust solutions that bridge the gap between impressive demos and reliable, production-ready tools, requiring both technical depth and an understanding of real-world user needs. You will be contributing to a mission-driven company focused on ensuring general-purpose artificial intelligence benefits all of humanity.
This role is based on-site in New York, NY, allowing for direct collaboration with your team and access to OpenAI’s vibrant work culture.
🎯 Application & Resume Tips for OpenAI
Craft your application to demonstrate a clear understanding of applied AI engineering within a production context. Focus on how your past experiences align with building dependable tools from research concepts.
Highlight Production Impact: Emphasize any projects where you’ve transitioned AI models from experimental stages to stable, performant production systems. Quantify your impact on metrics like latency, reliability, or cost if possible.
Showcase Agent Development: Detail your experience with building and optimizing AI agents, specifically mentioning how you’ve improved their functionality, usefulness, or steerability for specific tasks. Projects involving turning models into “useful behavior” are particularly relevant.
Emphasize Collaborative Skills: Since this role involves working with research, infrastructure, and product teams, provide examples of successful cross-functional collaboration where you contributed to a shared goal or integrated different technical components.
📎 How to Apply
Review the requirements above and ensure your resume reflects the key skills highlighted in this role. Applications are reviewed on a rolling basis, so applying early is strongly recommended.
👉 Click here to apply on the official OpenAI careers page.
