Principal Software Engineer, Agent Harness Bridge
OpenAI3 days ago
Location
San Francisco
Type
Full Time
Salary
USD 347,000 – 490,000
Level
Principal
Role
Principal Engineer
Posted
Jul 9, 2026
Full TimePrincipal
The role
Summary
Lead the architecture and evolution of OpenAI's Agent Harness Bridge, a critical integration layer connecting the Codex harness with training infrastructure. This Principal-level role combines backend infrastructure expertise with cross-functional leadership, requiring strong systems design and API architecture skills to support large-scale training workloads while maintaining engineering quality and research enablement at the frontier of AI systems.
What you'll do
Design and Evolve the Agent Harness Bridge Architecture: Architect and lead the evolution of the integration layer between the Codex harness and research training infrastructure, ensuring alignment with product runtime capabilities and training system requirements. Make principled technical decisions about abstraction layers, API contracts, and system boundaries that serve both immediate training needs and long-term platform evolution.
Own End-to-End Integration Surfaces: Take ownership of major integration points from initial architecture design through API specification, implementation, deployment orchestration, operational monitoring, and long-term maintenance. Establish clear accountability for system correctness, performance characteristics, and user experience across the research-to-infrastructure boundary.
Build Reliable Containerized Execution Systems at Scale: Design and implement containerized runtime systems capable of executing demanding training workloads reliably at scale. Focus on resource isolation, failure handling, observability instrumentation, and graceful degradation to support the unpredictable resource demands and failure modes of large-scale distributed training.
Lead Cross-Functional Technical Partnerships: Collaborate closely with research teams, agent systems developers, infrastructure engineers, and platform teams to understand emerging training use cases and evolving harness capabilities. Translate technical requirements into clean system designs while maintaining strong engineering discipline and preventing technical debt accumulation.
Design Ergonomic Interfaces for Technical Users: Create clean, stable interfaces and workflows that enable highly technical internal users to move quickly without sacrificing correctness or operational safety. Apply developer experience principles to research platform design, anticipating user workflows and reducing friction in the training infrastructure feedback loop.
Establish Durable Abstractions and Prevent Technical Debt: Build generalizable, well-documented abstractions that handle common patterns and edge cases, preventing one-off workarounds from calcifying into long-term technical debt. Establish clear ownership models and runbooks that enable other engineers to maintain and evolve systems reliably.
Raise Engineering Standards and Operational Rigor: Elevate standards for correctness, reliability, operational maturity, and engineering judgment across this critical research-facing system. Establish best practices for error handling, monitoring, incident response, and system design that model excellence for the broader infrastructure organization.
What we look for
Technical
Backend Systems Design and ImplementationAdvanced expertise in designing and building scalable backend systems, including experience with distributed systems concepts, service architectures, and integration patterns. Demonstrated ability to make principled architectural trade-offs and establish patterns that scale across team and organizational boundaries.
API Design and Contract DefinitionDeep proficiency in designing clean, stable, and well-documented APIs that serve both internal and external consumers. Strong experience with versioning strategies, backwards compatibility, and API evolution in fast-moving environments where breaking changes have organizational impact.
Python ProficiencyExpert-level Python development skills with experience in backend platform engineering, systems scripting, and infrastructure automation. Comfortable writing production-grade Python at scale with attention to code maintainability, performance, and testing strategies.
Infrastructure and Platform EngineeringStrong background in platform engineering, including containerization technologies (Docker, Kubernetes), infrastructure orchestration, and operational tooling. Experience building or maintaining internal developer platforms that serve multiple engineering teams or research groups.
Systems Reliability and Operational ExcellenceDemonstrated expertise in designing reliable systems with rigorous attention to failure modes, observability, monitoring, and incident response. Experience establishing operational runbooks, alerts, and diagnostic tools that enable sustained reliability across complex systems.
Containerization and Runtime ExecutionHands-on experience with containerized execution environments, container orchestration, resource management, and execution runtime design. Understanding of sandbox isolation, process management, and the operational challenges of supporting demanding workloads at scale.
Education
Bachelor's Degree in Computer Science or Related FieldBachelor's degree in Computer Science, Software Engineering, or related discipline. Equivalent experience demonstrating deep computer science fundamentals may be considered in lieu of formal degree.
Strong Foundation in Computer Science FundamentalsSolid understanding of core computer science concepts including algorithms, data structures, distributed systems theory, and operating systems principles. Applied knowledge of these fundamentals reflected in system design decisions and engineering practice.
Experience
Senior Backend or Infrastructure EngineeringMinimum 8+ years of backend or infrastructure engineering experience with demonstrated progression to senior levels. Track record of leading design and implementation of critical systems that required cross-functional coordination and delivered measurable business or research impact.
Leading Cross-Functional Technical InitiativesProven experience leading technical efforts that spanned multiple teams, engineering organizations, or different functional areas (research, infrastructure, platform). Demonstrated ability to build consensus, establish clear technical direction, and execute complex changes with organizational buy-in.
Building for Technical Internal UsersDirect experience building internal platforms, developer tools, or research infrastructure that serves demanding technical users. Understanding of developer experience principles, the importance of ergonomic design for internal systems, and how to gather and incorporate feedback from sophisticated users.
Scaling Systems in Fast-Moving EnvironmentsTrack record of growing and scaling backend systems and infrastructure within organizations experiencing rapid change. Experience managing technical debt, establishing engineering discipline while maintaining velocity, and making pragmatic decisions about when to refactor versus when to push forward.
Research-Facing Infrastructure or ML SystemsPrior experience working on or near research infrastructure, machine learning systems, or scientific computing platforms. Understanding of the unique demands researchers place on infrastructure and the trade-offs between flexibility and stability in research environments.
Skills
Required skills
PythonExpert-level Python development with production backend experience and familiarity with infrastructure automation and systems programming patterns.
API Design and Contract DefinitionPrincipled API design with deep understanding of versioning, backwards compatibility, and designing for scale and evolution.
Distributed Systems DesignStrong grasp of distributed systems concepts, service architecture patterns, and the operational challenges of scaling complex systems.
Systems Reliability and ObservabilityExpertise in designing reliable systems with comprehensive monitoring, alerting, and incident response infrastructure.
Containerization and OrchestrationHands-on proficiency with Docker, container runtime systems, and orchestration platforms for managing containerized workloads.
Cross-Functional Leadership and CommunicationAbility to lead technical discussions across diverse stakeholder groups, establish clarity on complex technical decisions, and drive consensus on architecture and implementation strategies.
Nice to have
RustExperience with Rust for systems programming, particularly in performance-critical infrastructure components or where memory safety and concurrency guarantees provide architectural advantages.
Kubernetes Administration and DesignDeep experience operating and designing systems on Kubernetes, including custom resource definitions, operators, and scaling strategies for complex workloads.
Machine Learning InfrastructureFamiliarity with ML training infrastructure, distributed training frameworks, or the specific challenges of supporting machine learning workloads at scale.
Research Infrastructure ExperiencePrior work on or adjacent to research platforms, scientific computing systems, or infrastructure designed to support experimental workflows.
gRPC and Protocol BuffersExperience designing and implementing gRPC-based service architectures with Protocol Buffers, particularly in performance-sensitive or cross-language environments.
Async/Concurrent Programming PatternsDeep experience with async programming models, concurrency primitives, and patterns for managing concurrent execution in Python or other languages.
Infrastructure as CodeProficiency with Infrastructure as Code tools and practices for managing complex infrastructure declaratively at scale.
Compensation & benefits
Salary
USD 347,000 – 490,000 (annual)
Stock options
Available
Benefits
Comprehensive Health Insurance
Medical, dental, and vision coverage with competitive plan options. OpenAI typically offers employer-covered premiums for employees and their dependents.
Equity Compensation
Competitive stock options or equity grants that align individual success with company growth, standard at principal engineering levels in high-growth AI companies.
Retirement Benefits
401(k) plan with employer matching contributions to support long-term financial security and retirement planning.
Unlimited Paid Time Off
Flexible vacation policy with no explicit limit, reflecting OpenAI's trust-based culture and work-life balance commitment.
Professional Development and Learning
Support for continued learning through conferences, training programs, and technical development aligned with career growth at the principal level.
Parental Leave
Comprehensive paid parental leave policies supporting both primary and secondary caregivers.
Mental Health and Wellness Support
Counseling services, mental health resources, and wellness programs to support employee wellbeing.
Commuter and Relocation Assistance
Support for Bay Area commuting and relocation assistance for those relocating to San Francisco for the role.
Interview process
- 1Initial Recruiter Screening — Phone or video conversation with OpenAI recruiter to assess background, experience level, motivation for the role, and general fit with OpenAI culture. Expect discussion of your infrastructure engineering background and leadership experience.
- 2Technical Systems Design Interview — Depth technical interview focusing on systems architecture, API design principles, and infrastructure patterns. You will likely discuss past systems you have designed, trade-offs you made, and how you would approach designing the agent harness bridge architecture.
- 3Backend Engineering Deep Dive — Detailed technical conversation with engineers on the infrastructure team covering Python proficiency, distributed systems knowledge, containerization experience, and approach to building reliable systems at scale. May include a design exercise.
- 4Cross-Functional Collaboration Discussion — Conversation with cross-functional stakeholders (research team members, platform engineers, other infrastructure leads) to assess your ability to work effectively across teams, understand diverse needs, and communicate technical concepts to different audiences.
- 5Leadership and Judgment Interview — Interview with senior engineering leadership assessing technical judgment, decision-making approach, experience mentoring and leading teams, and ability to establish standards and prevent technical debt in complex environments.
- 6On-site or Final Round — Potential on-site interview in San Francisco with multiple team members for final assessment of cultural fit, communication style, and ability to thrive in OpenAI's fast-moving research environment.
Apply for this position
You'll be redirected to the company's application page
More Jobs at OpenAI
84 other open positions
OpenAI
View all jobs
OpenAI is an American artificial intelligence research organization developing advanced AI models like GPT. Focused on ensuring AI benefits humanity, it creates tools for natural language processing and generative AI applications.
San Francisco, California, United StatesFounded 2015openai.com
Tech Stack
Languages
PythonRust
Frameworks
FastAPI or similar Python async frameworksgRPC
Databases
PostgreSQLRedis or similar caching layer
Tools
DockerKubernetesPrometheus and GrafanaGit and version control
Other
Container Runtime SystemsDistributed Training ConceptsResearch Platform Design
Interview Guides
5 guides available for OpenAI
Apply Now