← Dow Interview Insights

Dow·Software Engineer·Technical Phone Screen·Senior

Senior
Apr 2026

Summary

Deep technical interview at Dow for a Research Engineer role, focused almost entirely on SHAP analysis applied to nonlinear physical descriptors like SOAP. The interviewer clearly wanted to see whether you understood the method well enough to criticize it, not just use it.

Questions Asked (3)

Q1

Does SHAP's additive consistency property still hold in a meaningful way when applied to highly nonlinear force-field descriptors like SOAP?

Technical Trade-offsRoot Cause Analysis
Author's notes

This is where I started to feel the pressure.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Clarify that SHAP's additive consistency is a mathematical property of the explanation model, not the underlying function, so it still holds formally for any model including those using SOAP descriptors. Then discuss the practical meaningfulness: highly nonlinear descriptors can cause SHAP values to be unstable or misleading due to feature correlations and interactions, requiring careful interpretation and validation.

Pro tip: Acknowledge that while the math guarantees consistency, the real question is whether the explanations are actionable for domain scientists—so emphasize the need to validate SHAP values against known physics or simpler baselines before trusting them in decision-making.

1. Define the property

State that SHAP's additive consistency means if a model changes such that a feature's marginal contribution increases, its SHAP value does not decrease. This is guaranteed by the Shapley value axioms.

2. Separate math from meaning

Explain that the property holds mathematically for any function, including those with SOAP descriptors, because SHAP operates on the model's output. However, meaningfulness depends on whether the explanations reflect true feature importance in a chemically interpretable way.

3. Identify challenges with SOAP

Discuss how SOAP's high-dimensional, nonlinear, and correlated features can lead to issues like feature interaction dominance, unstable SHAP values across similar structures, and difficulty in attributing importance to individual components.

4. Propose validation strategies

Suggest using domain knowledge, synthetic tests, or comparing SHAP values with simpler descriptors to assess if the explanations are physically meaningful. Mention techniques like SHAP interaction values or clustering to handle correlations.

5. Conclude with trade-offs

Summarize that while consistency holds, the practical utility of SHAP for SOAP depends on the specific application and requires careful interpretation, possibly combining with other interpretability methods.

Key Points to Mention

  • Shapley value axioms: local accuracy, missingness, consistency
  • SOAP descriptors: high-dimensional, nonlinear, invariant to rotations/translations
  • Feature correlation and interaction effects in SHAP
  • Computational cost and approximation methods (e.g., KernelSHAP, DeepSHAP)
  • Domain-specific validation: comparing with physical intuition or simpler models
  • Alternatives: permutation importance, partial dependence plots, or integrated gradients

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

When using SHAP on highly nonlinear descriptors like SOAP, can it introduce overfitting or feature-redundancy artifacts, and why?

Technical Trade-offsRoot Cause Analysis
Author's notes

Blanked for a second here.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Clarify that SHAP itself is a model-agnostic explanation method and does not cause overfitting, but it can reveal or amplify artifacts when applied to highly nonlinear descriptors like SOAP. Discuss how correlated features and nonlinear interactions can lead to misleading SHAP values, and suggest mitigation strategies such as feature clustering or using conditional SHAP variants.

Pro tip: Emphasize that SHAP explanations are only as reliable as the underlying model and data; always validate interpretability with domain knowledge and consider using multiple explanation methods to cross-check.

1. Clarify SHAP's role

Explain that SHAP is an explanation technique, not a modeling technique, so it does not directly cause overfitting. However, it can expose or exacerbate issues in the model or data.

2. Address feature redundancy

Discuss how highly correlated features (common with SOAP descriptors) can lead to redundant SHAP values, where importance is split arbitrarily among correlated features, making interpretation misleading.

3. Discuss nonlinearity and interactions

Explain that SHAP assumes feature independence in its original form; with nonlinear interactions, it may attribute importance incorrectly, creating artifacts that look like overfitting but are actually misinterpretations.

4. Propose mitigation strategies

Suggest using conditional SHAP (e.g., TreeSHAP with conditional expectations), clustering features, or applying dimensionality reduction before SHAP to reduce redundancy and improve reliability.

5. Validate with domain knowledge

Stress the importance of validating SHAP results against domain expertise and using complementary methods like permutation importance or LIME to ensure robust interpretation.

Key Points to Mention

  • SHAP is model-agnostic and explains predictions, not a cause of overfitting.
  • Correlated features can lead to redundant and misleading SHAP values.
  • Nonlinear interactions can violate SHAP's independence assumption, causing artifacts.
  • Conditional SHAP variants (e.g., TreeSHAP) can handle feature dependence better.
  • Feature clustering or dimensionality reduction can mitigate redundancy.
  • Cross-validation with domain knowledge and other interpretability methods is crucial.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

What are the fundamental limitations of SHAP as a method, and how would you go beyond just using the library to improve attribution quality for nonlinear physical descriptors?

Technical Trade-offsSystem Design
Author's notes

Best part of the interview for me.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Acknowledge SHAP's theoretical limitations (e.g., reliance on feature independence, computational cost, and local linearity) and then propose enhancements tailored to nonlinear physical descriptors, such as using domain-specific kernels, interaction-aware methods, or hybrid approaches. Emphasize the importance of validating attributions against physical laws and expert knowledge.

Pro tip: Demonstrate awareness that SHAP is a tool, not a solution—show how you would combine it with domain knowledge and alternative methods like integrated gradients or causal inference to build trust in attributions.

1. Identify SHAP's core limitations

Discuss issues like the independence assumption, computational complexity for large feature sets, and the challenge of capturing nonlinear interactions. Mention that SHAP values can be misleading when features are correlated or when the model is highly nonlinear.

2. Relate limitations to physical descriptors

Explain how nonlinear physical descriptors (e.g., temperature, pressure, composition) often violate SHAP's assumptions due to correlations and complex interactions. Highlight that standard SHAP may not reflect true physical causality.

3. Propose enhancements beyond the library

Suggest methods such as using SHAP with custom background distributions, incorporating interaction indices (e.g., SHAP interaction values), or employing alternative attribution techniques like Integrated Gradients or LIME with domain-specific modifications.

4. Validate attributions with domain knowledge

Describe how to validate SHAP results against physical laws, simulation data, or expert intuition. Mention techniques like sensitivity analysis or comparing with known mechanistic models.

5. Discuss implementation and trade-offs

Talk about practical considerations: computational cost, scalability, and the need for robust software engineering practices. Emphasize that improving attribution quality may require custom code and iterative refinement.

Key Points to Mention

  • SHAP's assumption of feature independence and its limitations with correlated features
  • Computational complexity of SHAP, especially for high-dimensional data
  • The challenge of capturing nonlinear interactions and higher-order effects
  • Alternative attribution methods like Integrated Gradients, LIME, or causal inference
  • Domain-specific validation of attributions against physical laws and expert knowledge
  • The importance of custom implementations and software engineering for scalability

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.