Netflix LLM Serving: Triton, vLLM, and Production Tradeoffs
Posted on November 16, 2026
Agent Architecture Patterns State and Recovery
Posted on February 16, 2026
Evaluating AI Coding Assistants with Reproducible Tasks
Posted on November 29, 2025
AgentCore Runtime Identity Memory and Observability
Posted on October 21, 2025
Prompt Templates with Inputs and Evaluation Cases
Posted on September 3, 2025
Comparing LlamaIndex and LangChain on Concrete Tasks
Posted on August 5, 2025
Bedrock Knowledge Bases and Reproducible RAG Evaluation
Posted on July 10, 2025
Exchange Design Sequencing and Recovery
Posted on June 16, 2025
Serving Copilot Completions Through a Proxy
Posted on May 9, 2025
Spring AI and LangChain4j at the Same Application Boundaries
Posted on April 25, 2025
DeepSeek Models Distillation and Infrastructure Projects
Posted on March 1, 2025
Streamlit Reruns Forms and Session State
Posted on February 15, 2025
Kafka Integration Testing with Explicit Lifecycles
Posted on January 27, 2025
KEDA Activation HPA Scaling and Queue Backlogs
Posted on December 11, 2024
Haystack Retrieval with a Versioned Local Example
Posted on November 29, 2024
Quantiacs Stock Data and xarray Dimensions
Posted on October 11, 2024
LLaMA-Factory Fine-Tuning and Adapter Lifecycle
Posted on September 16, 2024
Powertools for Lambda with Logging Tracing and Metrics
Posted on August 22, 2024
RAG Inference Boundaries and Deployment Tradeoffs
Posted on July 3, 2024
A Brag Document with Evidence and Shared Credit
Posted on June 23, 2024
Pinterest Ad Retrieval Ranking and Model Layers
Posted on June 7, 2024
AST and LLM Test Migration at Slack
Posted on May 29, 2024
Portfolio Analysis at Scale and Computational Graphs
Posted on May 15, 2024
Chronon and the Training Serving Feature Contract
Posted on April 25, 2024
Performance Engineering as a Measurement Loop
Posted on April 3, 2024
Streaming Data Abstractions at Yelp
Posted on March 23, 2024
AWS Lambda Cold Starts and the Invocation Path
Posted on March 16, 2024
Distributed SQL Engines and Data Access
Posted on February 29, 2024
RAG Ingestion Retrieval and Answer Evaluation
Posted on February 8, 2024
Generally AI Notes on Language Models and Audio
Posted on February 3, 2024
Griffin Training Serving and Feature Contracts
Posted on January 7, 2024
How Staff Engineers Help Before During and After Incidents
Posted on January 6, 2024
Cloudflare ML Platform and GitOps Boundaries
Posted on December 9, 2023
Airflow and Flyte for Batch Workflows
Posted on November 21, 2023
Staff Engineering Technical Direction and Influence
Posted on November 19, 2023
Venice Within LinkedIn AI Infrastructure
Posted on August 9, 2023
People Search Retrieval and Ranking at LinkedIn
Posted on July 12, 2023
Lucene Inside Search Servers
Posted on May 30, 2023
Cloud Cost Optimization Through Workload Accounting
Posted on March 12, 2023
Jetson Nano Object Detection and Custom Actions
Posted on January 21, 2023
Explaining Why Before How and What
Posted on September 10, 2022
Comparing Cloud Edge Reconciliation Designs
Posted on June 22, 2022
Local Kubernetes Distributions and Container Tools
Posted on February 21, 2022
An ML Infrastructure Responsibility Map
Posted on December 29, 2021
Career Growth Coaching and Role Changes
Posted on November 7, 2021
SageMaker Pipelines and MLflow Solve Different Parts
Posted on June 19, 2021
Choosing Data Tools by Responsibility
Posted on May 3, 2021
SageMakerEstimator and the Spark Job Boundary
Posted on April 16, 2021
Dagster Plans Runs and Execution Boundaries
Posted on March 9, 2021
Lakehouse Storage Metadata and Query Execution
Posted on February 22, 2021
Questions Before Shipping a Model
Posted on September 19, 2020
Career Development Practices I Can Repeat
Posted on September 5, 2020
Slack Infrastructure and Message Boundaries
Posted on July 7, 2020
Netflix Meson and Dynamic ML Workflows
Posted on May 18, 2020
Netflix Data Processing and Storage Evolution
Posted on May 17, 2020
Netflix Playback and Identity Boundaries
Posted on May 16, 2020
Selected AWS Service Notes from April 2020
Posted on April 30, 2020
A Reinforcement Learning Battlesnake Workflow
Posted on March 27, 2020
Behavioral Interviews and Constructive Feedback
Posted on March 25, 2020
NGINX Workers Events and Graceful Reloads
Posted on March 24, 2020
Michelangelo and the ML Production Lifecycle
Posted on March 12, 2020
A Conceptual Ride-Hailing System Design
Posted on January 5, 2020
Similarity-Based Predictions at Uber
Posted on September 28, 2018
Sizing a Historical Elasticsearch Service Workload
Posted on December 10, 2017
Scaling Data Quality at Netflix
Posted on November 11, 2017
Crowd Motifs for Automated Mobile Testing
Posted on October 7, 2017
Understanding Spark Executor Memory
Posted on October 3, 2017
An IoT Stack with Spark and Ignite
Posted on September 23, 2017
Spark Shuffle Tuning at Facebook
Posted on September 14, 2017
Spark RDDs DataFrames and Datasets
Posted on September 9, 2017
Topic Models in GoDaddy Customer Success
Posted on August 17, 2017
Watermarks and Late Data in Structured Streaming
Posted on August 10, 2017
Netflix Network Monitoring with Kinesis
Posted on August 8, 2017
Running Spark on Kubernetes
Posted on August 1, 2017
User Behavior Anomaly Detection at Salesforce
Posted on July 31, 2017
Intel BigDL and Distributed Training
Posted on July 31, 2017
Spark Sessions Through Livy at PayPal
Posted on July 15, 2017
Random Walks on Large Graphs with Spark at LinkedIn
Posted on July 3, 2017
Hive Bucketing and Spark Join Plans
Posted on June 26, 2017
Scaling ML Workloads with Spark at Krux
Posted on June 21, 2017
Flint and Time-Aware Joins
Posted on February 15, 2017
Multitenancy and Isolation Choices
Posted on February 14, 2017
Structured Streaming ETL from Files
Posted on February 6, 2017
Distributed Tracing at Uber
Posted on February 5, 2017
About This Engineering Notebook
Posted on February 4, 2017