🧠 AI & Machine Learning · Sequence
Real-time Model Prediction Sequence
How an app gets a live prediction: features are fetched from an online store, the model scores the request, and the result and inputs are logged for monitoring.
Drawing diagram…
What this diagram shows
- Online features are read in milliseconds from Redis
- A fallback rule answers if the model times out
- Every prediction is logged to watch for drift
Prompt used
Sequence for real-time ML inference: a checkout app asks the prediction API for a fraud score. The API reads customer features from an online feature store (Redis), calls the model server, and if it does not answer in 100 ms uses a rule-based fallback. It returns the score and logs inputs and output to Kafka for drift monitoring.
Mermaid code
sequenceDiagram
participant APP as Checkout App
participant API as Prediction API
participant FS as Online Feature Store
participant MS as Model Server
participant FB as Fallback Rules
participant K as Kafka Log
APP->>API: Score this payment
API->>FS: Get customer features
FS-->>API: Features
API->>MS: Predict
alt Answer within 100 ms
MS-->>API: Fraud score 0.12
else Timeout
API->>FB: Rule-based score
FB-->>API: Score 0.30
end
API-->>APP: Score and decision
API->>K: Log inputs and outputRelated templates
RAG Chatbot over Company Documents
A chatbot that answers staff questions from company documents: documents are split and indexed in a vector database, and each question pulls the most relevant passages before the language model writes an answer.
RAG Question Answering Sequence
What happens when a user asks the document chatbot a question: permission check, embedding, vector search, prompt building and a cited answer.
Document Ingestion for Vector Search
How documents become searchable passages for an AI assistant: extraction, cleaning, chunking, embedding and indexing, with changed files re-processed automatically.
AI Agent Tool-Use Workflow
How an AI agent completes a task by planning, calling tools, checking results and asking a person to approve risky actions before it finishes.
MLOps Model Training Pipeline
Where each part of an MLOps setup runs: feature store, training jobs on GPU nodes, experiment tracking, model registry and automated deployment to serving.
ML Model Lifecycle
The states a machine learning model moves through, from experiment to production to retirement, including rollback when live performance drops.