Data Scientist
Confirmed on the employer's own hiring board on Oct 16, 2026. First seen by Alion on Sep 30, 2026. AgilityIO scores B on the Alion truth index.
Company Overview
AgilityIO is a software development firm that designs and builds custom applications for startups and Fortune 500 companies. We are a global team of over 400 developers, QA engineers, project managers, and UX/UI designers with offices in New York City and Vietnam.
This is a flexible, contract-based role open to candidates in Latin America.
About the Role
We are seeking a Data Scientist to evaluate AI-generated data analyses on real datasets, stress-test the reasoning behind them, and write the judgments that teach the next generation of models what rigorous data work looks like. The newest models can spin up a full analysis in minutes - load the data, pick a method, produce charts and a confident conclusion. Someone has to check whether the statistics actually hold up, and that's this role.
Responsibilities
Evaluate AI-built analyses on real datasets: the method choice, the assumptions, and whether the conclusion actually follows from the numbers
Red-team the statistics to expose leakage, p-hacking, confounded comparisons, and confident nonsense before real users trust them
Write the better analysis when the model falls short: sound methodology, honest uncertainty, clear takeaways
Build, train, and evaluate machine learning models (classification, regression, clustering, forecasting), including feature engineering, cross-validation, and metric selection
Document methodology, assumptions, and limitations so that work can be reviewed and reproduced
Collaborate with engineers, product managers, and other team members, and turn open-ended questions into well-defined analytical problems
Required Qualifications
Professional, academic, or serious independent experience doing real data analysis in industry or research
Proficiency in Python and its data stack (pandas, NumPy, SciPy, scikit-learn, statsmodels, or equivalents) and in SQL
Experience working in Jupyter or similar notebook environments
Experience with data visualization tools and libraries (matplotlib, seaborn, Plotly, or BI tools such as Tableau or Power BI)
Working knowledge of machine learning fundamentals: model selection, overfitting, evaluation metrics, and validation strategies
Clear written and spoken English
No degree required
Preferred Qualifications
Background as a Data Scientist, Data Analyst, or Analytics Engineer
Experience reviewing, auditing, or red-teaming someone else's analysis or model output
Familiarity with common pitfalls in applied statistics: p-hacking, leakage, confounding, multiple comparisons
Familiarity with cloud platforms (AWS, GCP, or Azure), data warehouses (Snowflake, BigQuery, Redshift), and MLOps practices such as experiment tracking and model versioning
Experience with Git and collaborative, version-controlled workflows
Prior experience with AI/ML data annotation, evaluation, or model-training projects
What We Look For
Skepticism and curiosity: you ask whether a result is real before you ask whether it is exciting
Honesty about uncertainty: you communicate what the data can and cannot support
Strong written communication and attention to detail
Self-direction: you are comfortable working independently and asynchronously in a remote environment
What We Offer
Competitive salary and performance-based bonuses
Flexible remote work environment
Professional growth opportunities and mentorship
Engaging and collaborative team culture with cutting-edge projects
Ready to Make an Impact?
If you're passionate about rigorous data work and want to help shape how the next generation of AI models reason about data, we'd love to hear from you - send your resume to [email protected].