A Friendly Tour of Process Memory on Linux
Infrastructure
AI's Dial-Up Era
Inside the race to build agent-native databases
Current New Orleans 2025
Confluent Intelligence likely was the biggest announcement. It’s a combination of three products:
Inside NVIDIA GPUs: Anatomy of high performance matmul kernels - Aleksa Gordić
Iceberg CDC: Stream a Little Dream of Me
Corrosion · The Fly Blog
Inside Husky’s query engine: Real-time access to 100 trillion events | Datadog
Event-Hidden Architectures | Skip
https://arxiv.org/pdf/2509.00997
R2 SQL: a deep dive into our new distributed query engine
This Send/Sync Secret Separates Professional From Amateur Rust Developers
[2504.16736] A Survey of AI Agent Protocols
Supporting our AI overlords: Redesigning data systems to be Agent-first
The Case for an Iceberg-Native Database: Why Spark Jobs and Zero-Copy Kafka Won’t Cut It
Hummock: A Storage Engine Designed for Stream Processing - RisingWave: Real-Time Event Streaming Platform
OpenConnect: LinkedIn's Next-Generation AI Pipeline Ecosystem
chroma/rust/wal3/README.md at main · chroma-core/chroma
wal3 uses a cryptographic hash to create a setsum of the data in the log. This setsum is an
associative and commutative hash function that is used to verify the integrity of the log. Because
of the way the hash function is constructed, it is possible to compute a new setsum from an existing
setsum and the setsum of a new fragment. This allows us to get cryptographic-strength integrity
checking of the log. We go into this at length in the verifiability section below.
wal3: A Write-Ahead Log for Chroma, Build on Object Storage·|·Chroma
Understanding Apache Fluss — Jack Vanlightly
How big are our embeddings now and why?
Thoughts on (Amazonian) Leadership
Replacing a cache service with a database - blag
MCP vs CLI: Benchmarking Tools for Coding Agents
Linearizability testing S2 with deterministic simulation
The Fork in the Road Post GPT-5
Evolving our real-time timeseries storage again: Built in Rust for performance at scale | Datadog
Iceberg Topics for Apache Kafka®: Zero ETL, Zero Copy
An Engineer's Guide to AI Code Model Evals
LLMs as Parts of Systems - Marc's Blog