You are helping me build the following AI agent workflow. ## Goal Survey Insights with Qdrant and AI: Imports survey responses to vector storage, clusters them by similarity, and summarizes key insights back into your spreadsheet. ## Specification - What it does: Import your Google Sheets survey responses to automatically group similar answers and summarize key insights directly back into your spreadsheet. This workflow uses AI to find patterns across your data so you can quickly understand the main takeaways without manual analysis. - Trigger: Run manually (Manual ยท on demand) - Autonomy: Fully hands-off - Expected setup effort: about an afternoon - Tools/services involved: - Google Sheets: Importing survey data and saving insights - Qdrant Vector Store: Storing vector embeddings for clustering - Embeddings OpenAI: Generating text embeddings - OpenAI Chat Model: Summarizing clustered responses and generating insights - Code: Running clustering algorithm via Python ## Known pitfalls, handle each one explicitly in your implementation 1. Qdrant collections require a predefined vector size that must match your embedding model output, so verify dimensions before inserting data. 2. Clustering high-dimensional text vectors often yields poor results without dimensionality reduction like PCA, so add a preprocessing step to reduce noise. ## Reference implementation https://n8n.io/workflows/2372 (workflow template) Fetch and inspect this before building. If it matches my stack, adapt it; if not, rebuild the pattern with my tools. ## Process requirements 1. Before building: ask me which of the listed tools I actually use and what my platform is (n8n / Make / code / other). Do not assume. 2. Adapt the pattern to my answers; do not force the reference stack. 3. Address every pitfall above; tell me how you handled each. 4. Provide a test plan I can run before letting this touch real data. 5. Ask before any step that sends messages, modifies data, or spends money. Source: https://usecasesforagents.com/use-case/survey-insights-with-qdrant-and-ai/ via usecasesforagents.com