Back to the Index

Online Inference Pipeline

What is an online inference pipeline?

An online inference pipeline is a program that runs in a model deployment and returns predictions to a client using a model, downloaded and cached from a model registry, and features that are either on-demand or precomputed and retrieved from the feature store.

What are the key steps in an online inference pipeline?

An online application sends a prediction request to the model deployment that contains an inference pipeline that processes the input request, computes any on-demand features, retrieves any precomputed features from a feature store, applies any model-dependent transformations to the on-demand and precomputed features and builds a feature vector from them, before making a prediction on the model with that feature vector, and finally post-processes the prediction before it sends the results back to the client application.

architecture graph

Interested for more?

Does this content look outdated? If you are interested in helping us maintain this, feel free to contact us.
O
AI Lakehouse
AI Pipeline
Auto-regressive Models
AutoML
O
Backfill features
Backfill training data
Backpressure for feature stores
Batch Inference Pipeline
O
CI/CD for MLOps
Compound AI Systems
Context Window for LLMs
O
DAG Processing Model
Data Compatibility
Data Contract
Data Lakehouse
Data Leakage
Data Modeling
Data Partitioning
Data Pipelines
Data Quality
Data Transformation
Data Type (for features)
Data Validation (for features)
Data Warehouse
Data-Centric ML
Dimensional Modeling and Feature Stores
Downstream
Query Engines
O
ELT
ETL
Embedding
Encoding (for Features)
Entity
O
Feature
Feature Data
Feature Engineering
Feature Freshness
Feature Function
Feature Groups
Feature Logic
Feature Monitoring
Feature Pipeline
Feature Platform
Feature Reuse
Feature Selection
Feature Service
Feature Store
Feature Type
Feature Value
Feature Vector
Feature View
Filtering
Fine-Tuning LLMs
O
Generative AI
Gradient Accumulation
Grouped Query Attention
O
Hallucinations in LLMs
Hyperparameter
Hyperparameter Tuning
O
Idempotent Machine Learning Pipelines
In Context Learning (ICL)
Inference Data
Inference Logs
Inference Pipeline
Instruction Datasets for Fine-Tuning LLMs
O
LLM Code Interpreter
LLM Temperature
LLMOps
LLMs - Large Language Models
Lagged features
LangChain
Latent Space
O
ML
ML Artifacts (ML Assets)
MLOps
MLOps Platform
MVPS
Machine Learning Infrastructure
Machine Learning Logs
Machine Learning Observability
Machine Learning Pipeline
Machine Learning Systems
Model Architecture
Model Bias
Model Deployment
Model Development
Model Evaluation (Model Validation)
Model Governance
Model Inference
Model Interpretability
Model Monitoring
Model Performance
O
Natural Language Processing (NLP)
O
Offline Store
On-Demand Features
On-Demand Transformation
Online Store
Online-Offline Feature Skew
Online-Offline Feature Store Consistency
Orchestration
O
KServe
PagedAttention
Pandas UDF
Parameter-Efficient Fine-Tuning (PEFT) of LLMs
Point-in-Time Correct Joins
Precomputed Features
Prompt Engineering
Prompt Store
Prompt Tuning
Python UDF
O
RLHF - Reinforcement Learning from Human Feedback
Real-Time Machine Learning
Recommender System
Representation Learning
Retrieval Augmented Generation (RAG) for LLMs
RoPE Scaling
O
SQL UDF in Python
Sample Packing
Schema
Similarity Search
Skew
Splitting Training Data
Streaming Feature Pipeline
Streaming Inference Pipeline
O
Test Set
Theory-of-Mind Tasks
Time travel (for features)
Train (Training) Set
Training Data
Training Pipeline
Training-Inference Skew
Transformation
Two-Tower Embedding Model
Types of Machine Learning
O
Upstream
O
Validation Set
Vector Database
Versioning (of ML Artifacts)
vLLM