New opportunity

Scientific Computing Engineer - Drug Product Process Modeling & Data Science

Novartis · Hyderabad, India

About this role

Novartis seeks a hands-on Scientific Computing Engineer in Hyderabad (Office), India, to strengthen process modeling, statistics and data science in drug product development. Working with formulation scientists, process engineers, data scientists and manufacturing experts, the engineer will build models and analytical tools that improve process understanding and support decisions from laboratory studies through scale-up. The position is full-time and regular.

Responsibilities

Develop mechanistic, empirical, statistical and hybrid models for formulation and process development, including manufacturing-relevant questions. Translate scientific questions into model- and data-ready problems with defined success criteria, assumptions and uncertainty considerations.
Use statistics, Design of Experiments, multivariate analysis and data-driven modeling to plan experiments and interpret results. Build predictive models and decision-support tools for drug product unit operations, particularly oral solid dosage forms, powder technology and process engineering.
Deliver reproducible, transparent data science solutions spanning data preparation, exploratory analysis, modeling, validation, deployment and lifecycle management. Create visualizations, dashboards and technical narratives for stakeholders; contribute to AI-assisted automation, documentation, internal standards and reusable modeling assets while maintaining scientific oversight.

Qualifications

A master’s degree or PhD in mechanical, process, chemical or pharmaceutical engineering, materials science, applied mathematics, statistics, data science or a closely related quantitative engineering discipline. An early-career profile is preferred: typically 2–4 years of relevant industry experience after a master’s degree, or 0–4 years after a PhD.
Strong engineering and mathematical foundations, such as process science, transport phenomena, statistics, numerical methods or mechanistic modeling. Hands-on programming experience in Python or a similar language is required, with the ability to become productive in Python quickly. Experience applying statistics, experimental design, analysis, simulation, optimization or machine learning to scientific or engineering problems is expected, alongside work with experimental and industrial datasets, uncertainty-aware interpretation and model credibility assessments against regulatory guidelines and standards. Strong communication skills are essential.

Beneficial or desirable experience includes machine learning and model evaluation, AI-assisted or agent-based workflows, model lifecycle and deployment considerations in regulated environments, dashboards, powder technology, formulation science, pharmaceutical unit operations, process modeling tools, PBM, DEM, gPROMS, digital twins, QbD, PAT and work in global matrix organizations. AI or machine learning experience is welcome but is not the primary selection criterion; these capabilities can be developed on the job.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonProcess modelingMechanistic modelingStatistical modelingData scienceStatisticsDesign of ExperimentsMultivariate analysisPredictive modelingData analysisData cleaningExploratory data analysisSimulationOptimizationMachine learningModel evaluationModel validationModel deploymentModel lifecycle managementData visualizationDashboardsAutomationNumerical methodsTransport phenomenaUncertainty analysisModel credibility assessmentPowder technologyFormulation scienceOral solid dosage formsPopulation balance modeling (PBM) computational methods (DEM)

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Frame useful modeling problems
The role translates formulation and process questions into model- and data-ready problems with subject-matter experts, defining success criteria, assumptions and uncertainty. Prepare to explain how a scientific question becomes a decision-focused analysis.
Model drug product processes
Develop mechanistic, empirical, statistical or hybrid models to improve process understanding and support scale-up and manufacturing-relevant decisions. Focus preparation on why a modeling approach fits a formulation or unit-operation question, particularly for oral solid dosage forms and powders.
Use experiments to accelerate learning
Apply Design of Experiments, statistics and multivariate analysis to plan studies and interpret results. Prepare to show how an analysis could guide the next experiment rather than merely describe existing data.
Deliver credible, reproducible tools
Responsibilities span data preparation through validation, deployment and lifecycle management, with transparency and reproducibility. The posting also calls for uncertainty-aware interpretation and model credibility assessments against regulatory guidelines and standards.
Make results usable across disciplines
Work with formulation, process, data science and manufacturing experts; communicate through visualizations, dashboards and technical narratives. The role also contributes documentation, reusable assets and AI-assisted workflows while retaining scientific oversight.

Likely challenges to think through

Balancing model detail with practical decisions
Likely preparation challenge, not a confirmed employer problem: a mechanistic model may aid interpretation but demand more data or assumptions than an empirical approach. Prepare to explain how you would compare approaches against the decision, available evidence, uncertainty and scale-up relevance.
Drawing defensible conclusions from imperfect data
Likely preparation challenge, not a confirmed employer problem: experimental and industrial datasets may require cleaning, while uncertainty limits what a predictive result can support. Prepare an example of checking data quality, testing model credibility and stating limitations plainly.
Turning analysis into a maintained workflow
Likely preparation challenge, not a confirmed employer problem: rapid analysis and AI-assisted automation must be balanced against reproducibility, practical usability and scientific oversight. Prepare to describe how you would document assumptions, validate outputs and communicate when human review is needed.

Practice your answers

Technical practice

How would you turn a formulation or process question into a modeling problem?

What to coverThe role calls for defining a model-ready question with subject-matter experts, including success criteria, assumptions and uncertainty. Suggested preparation: explain how you would identify the decision the model must support, the available data and what would count as a useful result.

When would you choose a mechanistic, empirical, statistical or hybrid model?

What to coverThe posting names all four approaches for drug product formulation and process development. Suggested preparation: compare their data needs, interpretability and limitations, and explain how the scientific question would guide your choice.

How would you assess whether a process model remains useful during scale-up?

What to coverThe role spans laboratory studies, scale-up and manufacturing-relevant questions, with engineering fundamentals such as transport phenomena. Suggested preparation: discuss which assumptions may change with scale and what evidence would test whether predictions remain credible.

How would you design and interpret experiments to improve process understanding?

What to coverThe posting calls for Design of Experiments, statistics and multivariate analysis to plan experiments and accelerate learning. Suggested preparation: describe how you would choose factors and responses, account for interactions and uncertainty, and use the results to decide what to study next.

How would you build a decision-support model for an oral solid dosage unit operation?

What to coverThe role highlights predictive tools for drug product unit operations, particularly oral solid dosage forms and powder technology. Suggested preparation: outline how you would define inputs and outputs with process experts, test predictive usefulness and communicate limits; do not assume a specific unit operation is required.

What would you check before modeling experimental and industrial datasets?

What to coverData cleaning, exploratory analysis and uncertainty-aware interpretation are stated core skills. Suggested preparation: cover data provenance, missing or inconsistent values, measurement variability and patterns that could make a model misleading.

How would you evaluate a model intended to support a process decision?

What to coverThe posting calls for validation and model credibility assessments against regulatory guidelines and standards. Suggested preparation: distinguish predictive performance from fitness for the intended use, and discuss uncertainty, validation data and documented limitations without assuming a particular regulatory framework.

How would you make a Python modeling workflow reproducible across its lifecycle?

What to coverHands-on programming in Python or a similar language is required, and the role covers preparation through deployment and lifecycle management. Suggested preparation: discuss traceable inputs, documented transformations, repeatable validation and how you would detect when a deployed model needs review.

How would you present model findings to stakeholders who are not modeling specialists?

What to coverThe role includes visualizations, dashboards and technical narratives that support decisions. Suggested preparation: describe how you would show the decision-relevant result, uncertainty and limitations without obscuring them behind model metrics.

Where could AI-assisted automation help a modeling workflow, and what should remain under scientific oversight?

What to coverThe posting includes AI-assisted work in data preparation, modeling, analysis and reporting, while treating AI experience as beneficial rather than primary. Suggested preparation: identify a task suitable for automation and explain how a person would check its outputs for scientific validity and practical usability.

Behavioural practice

Tell me about a time you clarified an ambiguous scientific or engineering question with others.

What to coverThe role requires turning formulation and process questions into defined problems with subject-matter experts. Suggested preparation: use a real situation/action/result example showing how you agreed on the objective, assumptions and a useful outcome.

Tell me about a time you explained a technical result to non-experts to support a decision.

What to coverClear communication and influencing decisions are stated expectations. Suggested preparation: use a real situation/action/result example that shows what you explained, how you conveyed uncertainty and what decision followed.

Tell me about a time unexpected data or model results changed your approach.

What to coverThe posting emphasizes hands-on analysis, data cleaning and uncertainty-aware interpretation. Suggested preparation: use a real situation/action/result example covering how you investigated the issue, revised your analysis and reported the outcome.

Tell me about a time you made analytical work easier for others to reuse.

What to coverThe role includes documentation, internal standards and reusable modeling assets within a global community. Suggested preparation: use a real situation/action/result example showing what you documented or shared and how others could use it.

Tell me about a time you learned an unfamiliar method while maintaining sound technical judgment.

What to coverThe posting values learning agility and scientific rigor, particularly for AI-assisted methods that can be developed on the job. Suggested preparation: use a real situation/action/result example showing how you learned, checked the method's suitability and recognized its limits.

Sharpen your knowledge

Topics to prepare

Engineering foundations for process models
Revision exercise: Sketch a simplified drug product unit operation, identify inputs, outputs, conservation assumptions and possible scale-up limitations. The posting calls for process science, transport phenomena, numerical methods and/or mechanistic modeling.
Choosing a modeling approach
Revision exercise: For a formulation question, compare when a mechanistic, empirical, statistical or hybrid model would be useful; state the assumptions and a measurable success criterion for each. These approaches and problem-framing steps are explicit accountabilities.
Experimental design and multivariate interpretation
Revision exercise: Outline a small Design of Experiments, then explain how you would interpret main effects, interactions and correlated measurements. The role applies statistics, DoE and multivariate analysis to plan experiments and accelerate learning.
Uncertainty and model credibility
Revision exercise: List sources of measurement, parameter and prediction uncertainty for a hypothetical process model, and describe evidence needed before using its predictions for a decision. The posting explicitly requires uncertainty-aware interpretation and credibility assessment against applicable regulatory guidelines and standards.
Drug product process context
Suggested preparation, not a core prerequisite: Trace how a change in powder properties might affect an oral solid dosage process and what data could test that explanation. The posting highlights these unit operations, while powder technology and formulation experience are listed as desirable.

Skills to revise

Program a reproducible analysis
Revision exercise: In Python, or a similar language you know, write a short analysis that loads example experimental data, calculates summary statistics and reruns from raw inputs. Hands-on programming is required, with quick productivity in Python expected.
Clean and explore process data
Revision exercise: On a practice dataset, flag missing values, inconsistent units and unusual observations; plot key relationships and record every cleaning decision. The role explicitly involves experimental and industrial datasets, data preparation and exploratory analysis.
Fit and validate a predictive model
Revision exercise: Fit a simple statistical or mechanistic model to example process measurements, check residuals and performance on held-out observations, and state where predictions would be unreliable. Predictive modeling, validation and uncertainty-aware interpretation are posting responsibilities.
Make an analysis usable across its lifecycle
Revision exercise: Package a small analysis with documented inputs, assumptions, validation results and instructions for rerunning and updating it. The posting calls for transparent, reproducible solutions spanning modeling, deployment and lifecycle management; this exercise does not imply a particular deployment platform.
Explain a modeling decision to non-experts
Revision exercise: Create one clear visualization and a brief technical narrative stating the question, result, uncertainty and recommended next experiment. The role requires communication that helps formulation, process and manufacturing stakeholders make decisions.

Your preparation checklist

Select a modeling project to walk through
The role calls for mechanistic, empirical, statistical, or hybrid models for drug product development. If available, prepare one scientific or engineering project walkthrough covering the question, model choice, assumptions, uncertainty, validation, and decision it informed.
Draft a process-modeling design exercise
As interview practice, sketch how you would turn a hypothetical oral-solid-dosage or powder-process question into a model-ready problem. Write down inputs, data needs, success criteria, scale-up considerations, and how you would check model credibility.
Prepare evidence of experimental analysis
The posting emphasizes statistics, Design of Experiments, multivariate analysis, and experimental or industrial data. Prepare a concise example showing data cleaning, experiment or analysis choices, results, and uncertainty-aware interpretation; use a hypothetical example if you cannot share prior work.
Make a reproducible coding example ready to explain
Hands-on programming in Python or a similar language is required. Prepare a small, shareable analysis or simulation with documented data preparation, clear code, validation checks, and a brief explanation of how another person could reproduce it.
Explain a model from analysis through use
The role spans exploratory analysis, modeling, validation, deployment, and lifecycle management. Outline how you would document assumptions, communicate limitations, monitor a model after use, and retain scientific oversight in any AI-assisted workflow.
Practice a decision-focused technical narrative
The posting calls for visualizations and explanations for non-experts. Prepare one figure or dashboard sketch and a two-minute explanation of what the evidence supports, what remains uncertain, and what decision it could inform.
Write questions for the interviewers
Ask which drug product unit operations need modeling most, how formulation and manufacturing experts define success, what data are available from laboratory studies through scale-up, and how model credibility and reusable assets are handled.

Availability last checked: 4 Nov 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…