Principal Machine Learning Architect, Enterprise Agentic Search (Remote)

You will be working on some of the following areas:

• Own the architecture for enterprise agentic search. Define how query understanding, source discovery, retrieval, agent orchestration, tool use, and context construction work together. Establish system boundaries, interfaces, and reusable capabilities that support multiple agent experiences.

• Design for enterprise complexity. Make architectural decisions for heterogeneous content, permissions, tenant isolation, freshness, domain-specific language, and information needs spanning multiple systems. Balance search quality and agent effectiveness with scalability, latency, cost, and reliability.

• Architect evaluation and experimentation. Define the benchmarks, datasets, evaluation environments, and observability needed to assess retrieval quality, agent search behavior, evidence coverage, and task success. Guide teams in building reproducible experiments and calibrated evaluators that inform architecture, model, and launch decisions.

• Set the direction for training LLMs for agentic search. Define model capabilities, learning objectives, training data requirements, and the architecture connecting training, evaluation, and serving. Lead technical decisions on supervised fine-tuning, preference optimization, and reinforcement learning to improve search planning, tool use, iterative evidence gathering, and grounded reasoning. Partner with engineers and scientists to train, validate, and deploy these improvements.

• Design the continuous improvement loop. Connect production interactions, agent trajectories, and failure analysis to evaluation cases, training data, and model updates. Establish data and feedback interfaces that preserve permissions, privacy, and evaluation integrity.

• Validate architecture through implementation. Build prototypes and reference implementations, investigate difficult system failures, and work directly with teams on critical components. Use experiments and production results to resolve trade-offs and guide incremental adoption.

• Lead technical direction across teams. Set a multi-quarter roadmap, align search, agent, and ML platform teams on shared architecture, and guide implementation through design reviews and technical mentorship. Take accountability for architectural outcomes and measurable improvements in agent task success.

What we are looking for

• Substantial hands-on industry experience architecting, delivering, and evolving production RAG or agentic search systems used by real customers, with particular depth in enterprise applications.

• A record of owning architecture from ambiguous customer needs through system design, implementation, launch, and evolution, with measurable improvements in search quality, task success, or customer outcomes.

• Demonstrated ability to define system boundaries, interfaces, and shared capabilities across teams, and to make sound architectural decisions about scalability, reliability, extensibility, and operational complexity.

• Practical experience with enterprise knowledge and workflows, including heterogeneous data sources, access permissions, freshness, domain-specific terminology, and complex information needs spanning multiple systems.

• Deep understanding of LLM-based search and agent systems, including retrieval, tool use, multi-step search, context construction, and grounded reasoning, and how these components affect end-to-end performance.

• Practical experience adapting and deploying LLMs, with the technical depth to guide training data design, supervised fine-tuning or other post-training methods, evaluation, and serving decisions. Experience improving search or tool-use behavior is especially valuable.

• Strong evaluation and experimental skills: you can design useful benchmarks, assess label and judge quality, recognize leakage and confounding factors, and connect offline results to product impact.

• Strong software and distributed-systems engineering skills, with the ability to validate designs in code and work deeply with data pipelines, model serving, and production observability.

• A record of influencing technical direction across multiple teams, guiding architecture through production adoption, and mentoring senior engineers and scientists.

• Sound judgment about trade-offs across model quality, data availability, latency, cost, reliability, privacy, and access control.

Benefits & Perks

Atlassian offers a wide range of perks and benefits designed to support you, your family and to help you engage with your local community. Our offerings include health and wellbeing resources, paid volunteer days, and so much more. To learn more, visit go.atlassian.com/perksandbenefits.

About Atlassian

At Atlassian, we're motivated by a common goal: to unleash the potential of every team. Our software products help teams all over the planet and our solutions are designed for all types of work. Team collaboration through our tools makes what may be impossible alone, possible together.

We believe that the unique contributions of all Atlassians create our success. To ensure that our products and culture continue to incorporate everyone's perspectives and experience, we never discriminate based on race, religion, national origin, gender identity or expression, sexual orientation, age, or marital, veteran, or disability status. All your information will be kept confidential according to EEO guidelines.

To provide you the best experience, we can support with accommodations or adjustments at any stage of the recruitment process. Simply inform our Recruitment team during your conversation with them.

To learn more about our culture and hiring process, visit go.atlassian.com/crh.

In line with local law, identity verification (which may include use of biometric data) is a condition of employment with Atlassian for employment fraud purposes.