Research Scientist - Information Theory and Statistical Inference

Evolver; Palo Alto (CA); On-site; USD 170,000 - 250,000; Full time

Job summary

Evolver is seeking a Research Scientist with a deep background in information theory, probabilistic modeling, and statistical inference to advance methods for transforming large-scale enterprise data into actionable information.

You will develop information-content metrics, quantify uncertainty as new evidence arrives, and design algorithms that guide data representation, graph structures, and decision-making in complex systems.

Qualifications

• Ph.D. or thesis-based Master’s degree in Electrical Engineering, Applied Mathematics, Statistics, Computer Science, Operations Research, Information Science, Signal Processing, or a related quantitative field.
• Strong foundation in graph theory demonstrated through a thesis, publications, or research projects.
• Strong mathematical foundation in probability, statistics, information theory, or statistical inference.
• Research experience in information theory, entropy and mutual information, Bayesian or statistical inference, uncertainty quantification, probabilistic modeling, or related areas.
• Experience with information bottleneck methods, rate-distortion theory, probabilistic graphical models, or Bayesian experimental design is a plus.

Responsibilities

• Develop methods to quantify information content, information gain, uncertainty, redundancy, and information loss across complex data systems.
• Decide how enterprise data should be represented as graphs and how those representations support reasoning.
• Develop approaches to determine which observations and data sources are most informative for downstream reasoning and decisions.
• Apply information theory, Bayesian inference, statistics, and probabilistic modeling to heterogeneous enterprise data.
• Study how information is preserved, compressed, combined, or lost as data is transformed into higher-level representations.
• Build algorithms, prototypes, benchmarks, and evaluation methods for information-aware AI systems.
• Collaborate with research, engineering, and product teams to translate new methodologies into real systems.

Skills

Graph theory

Probability & statistics

Information theory

Bayesian inference

Statistical inference

Python / MATLAB

Education

PhD or Master’s in a quantitative field

Tools

Python

MATLAB

Job description

Research Scientist - Information Theory and Statistical Inference

About Evolver

Evolver is an AI technology company transforming professional services. We combine deep domain expertise, advanced AI, and continuous applied learning to automate complex enterprise workflows while keeping people involved where judgment matters. Our goal is to help organizations operate more efficiently, accurately, and intelligently.

The Role

Evolver is looking for a Research Scientist with a strong background in information theory, statistical inference, uncertainty quantification, or related mathematical methods.

You will help develop methodologies for understanding how large-scale, heterogeneous enterprise data can be transformed into decision-relevant information, how uncertainty changes as new evidence isacquired, and how the value of information can be quantified.

We welcome applications from exceptional recent graduates as well as experienced researchers. We care more about mathematical depth, research ability, and original thinking than years of industry experience.

What You'll Do

• Develop methods to quantify information content, information gain, uncertainty, redundancy, and information loss across complex data systems.
• Part of this work is deciding how enterprise data should be represented as graphs, and how well those representations support reasoning.
• Develop mathematical approaches fordeterminingwhich observations and data sources are most informative for downstream reasoning and decisions.
• Apply information theory, Bayesian inference, statistics, and probabilistic modeling to heterogeneous enterprise data.
• Study how information is preserved, compressed, combined, or lost as data is transformed into higher-level representations.
• Develop methods for identifying information gaps and determining the value of acquiring additional evidence.
• Build algorithms, prototypes, benchmarks, and evaluation methods for information-aware AI systems.
• Collaborate with research, engineering, and product teams to translate new methodologies into real systems.

Minimum Qualifications

• Ph.D. or thesis-basedMaster's degree in Electrical Engineering, Applied Mathematics, Statistics, Computer Science, Operations Research, Information Science, Signal Processing, or a related quantitative field(Note: Please include the title of your thesisin your applicationand a brief summary of the problem, methodology, and your specific contribution).
• Strong foundation in graph theory, such as graph algorithms, combinatorial optimization, graph decomposition, or compact graph representations, demonstrated through a thesis, publications, or research projects.
• Strong mathematical foundation in probability, statistics, information theory, or statistical inference.
• Research experience in one or more of:
• information theory
• entropy and mutual information
• Bayesian or statistical inference
• uncertainty quantification
• probabilistic modeling
• signal processing
• representation learning
• active learning or experimental design
• Strong Python, MATLAB, or equivalent scientific computing skills.
• Ability to translate mathematical concepts into computational methods and working prototypes.

There is no minimum number of years of industry experience. Strong candidates maydemonstratetheir capabilities through a thesis, publications, research projects, internships, open-source work, or relevant industry experience.

Preferred Qualifications

• Experience with information bottleneck methods, sufficient statistics, rate-distortion theory, value of information, compressed sensing, Bayesian experimental design, or probabilistic graphical models.
• Experience working with large-scale, noisy, heterogeneous, or partiallyobserveddata.
• Research publications ordemonstratedresearch impact.
• Experience connecting theoretical methods to real-world data or decision-making problems.
• Familiarity with modern machine learning,representationlearning, foundation models, or agentic AI systems.

Deep prior experience with LLMs, RAG, prompt engineering, or specific agent frameworks is notrequired.

About Evolver

Evolver is building intelligent enterprise systems that transform large volumes of heterogeneous data into structured information for reasoning and decision-making.

We are looking for researchers who want to bring the foundations of information theory, statistical inference, and uncertainty into the next generation of AI.

As set forth in Evolver’s Equal Employment Opportunity policy,we do not discriminate on the basis of any protected group status under any applicable law.