Menu

#Embedding

63 posts

Feed·
20 of 63 posts
Laravel 13 Vector Search: Semantic Queries Without a Search Service
📰
0Read More

Laravel 13 Vector Search: Semantic Queries Without a Search Service

DEV Community: laravel·Danny Holloran·5 days ago
#dc9MUaQJ
#devtolaravel#devto#dev#embedding#search

Laravel 13 adds whereVectorSimilarTo, vector columns, and an AsVector cast, so meaning-based search can live in the Postgres or MariaDB database you run.

15s
We Built a CV Scanner That Compares Skills Against Real Jobs. Then Production Exposed a Bottleneck.
📰
0Read More

We Built a CV Scanner That Compares Skills Against Real Jobs. Then Production Exposed a Bottleneck.

DEV Community: backend·Derrick Muturi·10 days ago
#1BXxeBHC
#devtobackend#devto#dev#skills#embedding

Building an AI-powered application is one thing. Getting it to behave reliably when someone actually...

15s
GitHub Octoverse 2026: What 7.38 Billion Monthly Commits Reveal About Agent-Generated Code Infrastructure
🖼️
0Read More

GitHub Octoverse 2026: What 7.38 Billion Monthly Commits Reveal About Agent-Generated Code Infrastructure

DEV Community: github·mech.app·11 days ago
#FwGTUwCu

How GitHub measures agent activity at scale using LLM-based classification, VPN-adjusted geolocation, and embedding models across 30M new repos per month.

15s
Benchmarking Dify's Vector Store Layer: Qdrant vs pgvector Under 500 RPS Concurrent Agent Queries
🖼️
0Read More

Benchmarking Dify's Vector Store Layer: Qdrant vs pgvector Under 500 RPS Concurrent Agent Queries

DEV Community: database·Samuel Oakley·12 days ago
#zMB7U3e4
#devtodatabase#devto#dev#vector#embedding

Load testing Dify's RAG retrieval backends with wrk and pytest-asyncio. Measuring p99 latency, memory overhead, and recall drift at production scale.

15s
Why VoiceStudio's Semantic Memory Fails Under Load—and How We Fixed It
🖼️
0Read More

Why VoiceStudio's Semantic Memory Fails Under Load—and How We Fixed It

DEV Community: database·Samuel Oakley·17 days ago
#gFhKubOL
#devtodatabase#devto#dev#hnsw#embedding

HNSW index contention, embedding normalization bugs, and connection pool exhaustion—three production incidents that taught us to separate voice synthesis from vector writes.

15s
25 Million Records Later: Why Your RAG Pipeline Still Isn't Production-Ready
📰
0Read More

25 Million Records Later: Why Your RAG Pipeline Still Isn't Production-Ready

DEV Community: database·Sanu Khan·19 days ago
#dDwlQzKs
#devtodatabase#devto#dev#ingestion#retrieval

A practical architecture review of AWS's Ray-based RAG ingestion pipeline—and the reliability, security, and retrieval gates needed to operate RAG in production.

15s
Keeping Vector Retrieval Stable While Your Knowledge Base Is Still Loading
🖼️
0Read More

Keeping Vector Retrieval Stable While Your Knowledge Base Is Still Loading

DEV Community: database·Samuel Oakley·24 days ago
#01OTFfIG
#devtodatabase#devto#dev#ingestion#retrieval

Separate ingestion from retrieval, handle HNSW contention, and build a runnable PostgreSQL boundary that won't collapse under concurrent load.

15s
How Generative Recommenders Are Redefining RecSys at Scale
🖼️
0Read More

How Generative Recommenders Are Redefining RecSys at Scale

NVIDIA Technical Blog·Elizabeth Goodman·25 days ago
#gxkLYdck
#developer#embedding#recsys#semantic#cache

Recommender systems (RecSys) are one of the most ubiquitous machine learning problems in the consumer internet industry yet notoriously difficult to train and…

15s