Skip to main content

Functional Usage

Q: How do I connect my own data source like S3 or SharePoint in TestingAIde?​

A: You can add a connector by selecting the data source (S3/SharePoint) in the RAG configuration panel and providing access credentials (bucket path, API keys, etc.).

Q: What happens if my document is too large for the model?​

A: TestingAIde uses text splitters to break down documents into smaller chunks so that they fit within token limits without losing context.

Q: Where are my trained adapters stored?​

A: Adapters are stored in the cloud storage configured by the user. For example, a Data Scientist can use their own S3 bucket for persistence.

Q: Can I edit the generated test scripts after generation?​

A: Yes, scripts can be edited directly in the dock before executing or downloading.

Q: Do I need coding skills to generate and run test scripts?​

A: No. The process is designed to be UI-driven — select test cases, choose language/framework, and generate. However, advanced users can modify scripts if needed.

Q: What's the difference between RAG and fine-tuning, and when should I use each?​

A: Use RAG when your data is versatile (PDFs, docs, enterprise knowledge) and changing frequently.
Use fine-tuning when you have a stable, high-quality dataset and want a model to deeply specialize.

Q: What role do embeddings play in the RAG pipeline?​

A: Embeddings are vector representations of text. They allow semantic search by comparing query embeddings with document embeddings to fetch the most relevant chunks.

Q: Does TestingAIde support both OpenAI and Hugging Face embeddings?​

A: Yes. You can choose embeddings from providers like OpenAI (high accuracy, hosted service) or Hugging Face (open-source, customizable).

Q: What's the advantage of adapters over full fine-tuning?​

A: Adapters let you adapt models quickly and cheaply without retraining the entire model. They're lightweight and stored per user's cloud bucket.

Q: How does TestingAIde ensure execution results are reliable?​

A: Execution results are displayed immediately with logs, and historical trends are tracked on the dashboard for accuracy validation.

Q: How does TestingAIde reduce the overall test lifecycle time?​

A: By automating test case generation, script generation, and execution, while also connecting directly with product requirements (PRDs), TestingAIde reduces manual effort and accelerates delivery cycles.

Q: How can TestingAIde integrate with enterprise workflows?​

A: TestingAIde connects to enterprise data sources (S3, SharePoint), stores adapters in user-defined storage, and generates output scripts in enterprise-preferred languages and frameworks, ensuring seamless adoption.

Q: What business scenarios benefit most from RAG vs fine-tuning?​

A:

  • RAG: Customer support knowledge bases, compliance document Q&A, product manuals.
  • Fine-tuning: Domain-specific fraud detection, financial risk scoring, customized NLP pipelines.

Q: How does TestingAIde ensure security of data and credentials?​

A: User credentials (like Twilio tokens, S3 keys) are securely stored with role-based access controls. Adapters and generated assets remain in user-owned cloud storage.

Q: What insights can leadership teams gain from TestingAIde dashboards?​

A: Leaders can track execution trends, failure rates, productivity improvements, and alignment of testing with business goals — enabling data-driven decision making.