Applied Scientist

About the role

The role bridges applied research and production to build adaptive machine learning systems. In this position, you will own the design and delivery of systems that learn and adjust in real-world conditions, translating research insights into reliable production behavior. You will work at the intersection of algorithms, interface design, and deployment, ensuring that adaptive methods meet practical constraints and user needs. A core part of your responsibility will be to frame ambiguous problems, select appropriate methods, and iterate based on observed performance in dynamic environments. You will collaborate closely with engineers and business stakeholders to align technical solutions with real-time learning goals and operational requirements. Your work will directly influence how the company builds efficient intelligence that evolves rather than remaining static. Finally, you will contribute to a portfolio of analyses and models that demonstrate the impact of adaptive systems on business outcomes.

Key facts

What you'll do

Design and implement real-time learning systems that adapt to changing data distributions and user contexts through interface-aware deployment strategies.

Evaluate and compare gradient-free, adaptive, and efficient methods to select approaches that balance accuracy, latency, and resource usage in production.

Lead problem selection by analyzing opportunity spaces and defining evaluation criteria that focus effort on high-value, deployable research.

Partner with engineering teams to operationalize models, ensuring that data modeling and curation decisions are reflected in robust training and inference pipelines.

Apply fluency in software engineering and ML frameworks such as PyTorch, JAX, and TensorFlow to prototype, iterate, and scale adaptive algorithms.

Investigate online learning, reinforcement learning, and efficient architectures to support deployment in dynamic, resource-sensitive contexts.

Assess human feedback and reward signals as model refinement tools, integrating them into training regimes where appropriate.

Communicate technical findings and trade-offs to non-technical stakeholders, aligning model behavior with high-level business goals and constraints.

Drive ownership and execution by prioritizing actions, iterating quickly, and maintaining curiosity about system behavior in the wild.

Define and track metrics that capture model performance, user experience, and business impact across experiments and rollouts.

Participate in the full ML workflow, from data curation and hypothesis formation through deployment, monitoring, and continuous improvement.

Champion best practices in experiment design, logging, and reproducibility to ensure that adaptive methods can be understood and trusted.

Contribute to the Adaption Passport program by using the annual travel stipend to explore new environments, data sources, and learning scenarios.

Engage with the interview process, which may include SQL or coding exercises, statistics questions, case studies, and discussions of past project impact.

Bring a clean write-up of previous analyses to interviews, demonstrating how you communicate uncertainty, business impact, and decision quality.

Requirements

The posting states a bachelor's degree requirement. You must have three to four years of industry experience in machine learning or applied research, with a track record of deploying systems that solve business problems.

You must demonstrate fluency in software engineering and ML frameworks such as PyTorch, JAX, and TensorFlow to support implementation and iteration.

You must have experience with online learning, reinforcement learning, or efficient architectures to operate effectively in dynamic, resource-sensitive contexts.

You must understand data modeling and curation decisions so that training setups are correctly aligned with model performance outcomes.

You must communicate technical work to non-technical audiences, ensuring alignment across teams and clear translation of methods into decisions.

You must show ownership, curiosity, and a bias toward action to drive execution and problem solving in ambiguous settings.

Experience with human feedback, reward signals, or adaptive learning methods is a bonus for model refinement and system tuning.

Practical notes

An annual travel stipend supports exploration through an Adaption Passport program.

Typical interview steps

Data interviews commonly include a SQL or coding exercise, a statistics question, and a case study. Candidates may be asked to design a metric, interpret an experiment, or build a small model. Some companies give a take-home analysis. Expect questions about past projects and the business impact of your work. Interviewers often evaluate how you communicate uncertainty and business impact, not only the math. Bringing a clean write-up of a past analysis to the interview is well received.

Good to know

Applied Scientist roles blend research experimentation with delivery of working systems.

Machine learning workflows cycle quickly through modeling, data work, and deployment stages.

Efficient and online learning methods lower resource demands while maintaining accuracy in changing conditions.

Human feedback and reinforcement learning techniques adjust models to user intent and environmental shifts.

Data curation and modeling choices together determine how well systems learn from real-world information.

Questions to ask

Good questions to ask the employer in the interview: what does success look like in the first six months, how is the team structured, what is the current biggest challenge, and how are decisions made. Asking about growth paths and the review process is also well received. Employers expect questions, and good ones show preparation.

Career growth

Data careers grow toward senior analyst, staff data scientist, or data engineering lead. Many professionals specialize in machine learning, analytics, or infrastructure. Cross-functional work with product and engineering teams becomes more important at senior levels. The field changes quickly, so continuous learning is part of the job. Professionals who can translate numbers into decisions tend to advance fastest.

About the company

Most AI is frozen in place - it doesn't adapt to the world. We think that's backwards. Our mandate is to build efficient intelligence that evolves in real-time.