NSNate ShakouriPrincipal Scientist

Case Study

AI-Assisted Workflow For Catalyst Synthesis And Process Design

A public-safe case study summary. Proprietary details, exact operating conditions, and sensitive performance data are omitted unless already public or approved for release.

Case Study

Title

AI-Assisted Workflow For Catalyst Synthesis And Process Design

Technical Area

AI-assisted scientific R&D, catalyst synthesis, process design, and technical documentation

Problem

Catalyst and process design require synthesis of literature, experimental constraints, characterization data, process assumptions, safety requirements, and engineering judgment. Unstructured AI use can create traceability and validation risk.

Approach

Use human-in-the-loop AI workflows for literature triage, hypothesis formation, experiment planning, decision logging, technical writing, and review. The workflow is designed to support scientific judgment rather than replace experimental validation.

Methods Used

  • Structured literature extraction
  • Retrieval-supported technical review
  • Hypothesis and design-space mapping
  • Experiment-planning and DOE scaffolds
  • Decision logs and assumption tracking
  • Human review checkpoints
  • [Approved internal-tool description placeholder]

Outcome

Created a cautious workflow pattern for accelerating catalyst and process design tasks while preserving review, traceability, and domain-owner accountability. Internal prompts, data sources, and toolchains are not disclosed.

Technical Significance

The work positions AI as a disciplined research assistant for scientific R&D, especially where catalyst design, process constraints, and documentation must stay connected.

Confidentiality Note

Public summary only. Omits internal prompts, datasets, proprietary documents, employer systems, non-public experimental results, and partner-specific use cases.

Related Public Source

Profile context