Member of Technical Staff - Machine Learning Research Engineer I
- Palo Alto · CA
- Posted 8 h ago
Checked at the source 1 h ago
With your free Joonow account, you continue to the company site and the job stays in your applications.
No real employer charges a fee, a course, an exam or a uniform. Suspicious?
Job description
Member of Technical Staff - Machine Learning Research Engineer I
San Francisco Bay Area | Bellevue, Washington
About AIVista
NTT DATA AIVista, Inc., a wholly owned subsidiary of NTT DATA, is based in Silicon Valley. We develop AI products for enterprises operating in complex regulatory environments. We work with NTT DATA Inc., NTT DATA Japan, and NTT DOCOMO to deploy systems for NTT clients. Our clients combine AIVista's product AI expertise with NTT DATA's industry knowledge and systems-integration experience.
The Role
The Machine Learning Research Engineer works closely with AIVista scientists to turn promising research into dependable capabilities for enterprise use. You will develop a deep understanding of each method, its assumptions, supporting evidence, and known limitations, then focus on the work required to make it reproducible, robust, scalable, and maintainable.
You will test how methods behave across realistic workloads, varied domains and data, difficult edge cases, and operational constraints; improve their reliability and efficiency; and integrate them into product capabilities with the traceability, controls, and review boundaries required in regulated industries. This work spans reinforcement learning, simulation environments, formal methods, autoformalization, and AIVista's data, rule, action, and process ontologies.
Responsibilities
Work with scientists to understand each method's assumptions, supporting evidence, known failure modes, and acceptance criteria for broader use.
Reproduce research results across independent runs and settings, characterize sources of variance, and identify behavior that depends on a narrow dataset, model version, configuration, or operating condition.
Refactor and extend research prototypes without changing their scientific meaning; make tradeoffs explicit and validate each material change against agreed baselines.
Design stress tests, adversarial cases, counterfactual scenarios, rare events, and out-of-distribution evaluations that expose brittle behavior before product use.
Scale methods from research-sized examples to realistic workflow length, data volume, concurrency, and domain variation while controlling latency and cost.
Improve reliability, calibration, constraint adherence, recovery behavior, and consistency, and determine whether measured gains generalize beyond the original research cases.
Advance synthetic environments, workflow simulations, and agent evaluations from research prototypes into stable, versioned assets that represent exceptions, human review, governed actions, and external-system behavior.
Harden autoformalization and ontology-backed methods so source traceability, rule and process semantics, validation evidence, and governed actions remain correct as content and scale change.
Preserve independent checks and acceptance evidence as models, code, data, ontologies, and process definitions evolve.
Feed findings back to Science, including failures that require changes to the problem formulation, learning method, reward or constraint design, formalization approach, or evaluation strategy.
Partner with product and engineering teams to define operating limits, monitoring, regression tests, release criteria, rollback plans, and ownership for capabilities moving into production.
Build the durable software and automation required for reproducible experimentation, repeated evaluation, efficient scaling, and safe product integration, and maintain clear documentation for future contributors.
Qualifications
Basic
Master's degree in computer science, machine learning, artificial intelligence, software engineering, applied mathematics, or a related field.
At least 5 years of experience taking machine-learning or research prototypes toward reliable, sustained use at realistic scale.
Demonstrated experience improving the reproducibility, robustness, scalability, or operating efficiency of a research capability, supported by clear measurements and failure analysis.
Strong software-engineering judgment, including the ability to restructure prototypes, isolate assumptions, design effective tests, diagnose performance and reliability problems, and leave maintainable code and documentation.
Working knowledge of machine learning, statistical evaluation, experimental design, and failure analysis, with the ability to implement methods from papers and reason about how engineering changes affect the method.
Experience evaluating behavior across changing data, model versions, workloads, and operating conditions rather than relying on a single benchmark result.
Ability to collaborate closely with scientists, engineers, product teams, and domain experts to move research capabilities toward reliable use in regulated environments.
Preferred
Experience hardening reinforcement-learning environments, agent systems, workflow simulations, digital twins, or other sequential-decision methods for use beyond research demonstrations.
Experience evaluating long-horizon agents, verifier-guided methods, or reinforcement learning with verifiable rewards under adversarial, rare, or changing conditions.
Experience scaling ontology-backed, knowledge-graph, rule, process, or provenance-aware methods while preserving semantic correctness and source traceability.
Experience taking formal specifications, autoformalization, program analysis, or verification methods from research prototypes into repeatable software-development workflows.
Experience diagnosing distribution shift, specification gaming, reward hacking, data leakage, calibration failures, or regressions that appear only at larger scale.
Experience establishing operating limits, monitoring, release evidence, and safe change practices for machine-learning or decision systems.
Experience deploying AI or decision systems in regulated or high-stakes industries such as insurance, financial services, healthcare, or public-sector operations.
A record of technical impact through shipped systems, research publications, patents, benchmarks, open-source work, or widely used internal research capabilities.
Benefits
Medical, dental, and vision insurance
401(k) plan
Significant company Health Savings Account contribution
Paid holidays and flexible paid time off
Compensation
The estimated base salary range for this role is $300,000-$400,000 USD. The salary for the successful applicant will depend on job-related factors such as education, training, work experience, business needs, and market demands. This range may be modified in the future. Total compensation also includes variable pay in the form of an annual target bonus and other cash incentives, with a combined potential of up to 100% of base salary..
Equal Opportunity Employment
NTT DATA AIVista is an equal opportunity employer. We do not discriminate based on race, religion, color, national origin, ancestry, sex, gender identity or expression, sexual orientation, age, disability, genetic information, marital status, military or veteran status, reproductive health decisions, or any other characteristic protected under applicable federal, state, or local law.
Joonow analysis
Joonow figures for this job, calculated from the jobs open on November 22, 2026.
Member of Technical Staff salary in Palo Alto - CA
This job doesn’t list a salary. The median for the role is US$ 22,150 a month.
Half of the jobs pay between US$ 20,830 and US$ 25,110. Based on 20 open jobs that list a salary.
Other medians for the role: Mid / senior: US$ 22,630; Hybrid: US$ 20,420.
See the Member of Technical Staff salary guideHiring pace
Ntt Data Aivista has 9 open jobs on Joonow. 4 of them were posted in the last 7 days.
See the 9 jobs from Ntt Data AivistaWhat the job asks for
- Machine Learning
- Reinforcement Learning
- Simulation Environments
- Formal Methods
- Autoformalization
- Stress Testing
- Adversarial Cases
Skills Joonow read in the posting.