Miguel Ángel Pastor Olivar

Fitting a Query Engine in Three Cache Lines Why the bytecode that powers your analytical queries is smaller than a tweet, and why that matters more than you think.

Fitting a Query Engine in Three Cache Lines

Closing the verification loop, Part 2: Fully autonomous optimization | Datadog Learn how Datadog achieves fully autonomous, verified code optimization in production using LLM-driven evolution, formal verification, and live traffic validation.

Closing the verification loop, Part 2: Fully autonomous optimization

Closing the verification loop: Observability-driven harnesses for building with agents | Datadog Learn how Datadog verifies AI-generated systems at scale using deterministic testing, formal methods, and observability-driven feedback loops.

Closing the verification loop: Observability-driven harnesses for building with agents

turbopuffer: Object Storage-native Database for Search (Simon Eskildsen) CMU Database Group — PostgreSQL vs. The World Seminar Series (Spring 2026) Speakers: Simon Eskildsen (https://www.linkedin.com/in/sirupsen) March 09,…

turbopuffer: Object Storage-native Database for Search www.youtube.com/watch?v=pqoR...

MD-MVCC: Multi-version Concurrency Control for Schema
Changes in Azure SQL Database

Using TLA+ to Fix a Very Difficult glibc Bug - Malte Skarupke - C++Now 2025 https://www.cppnow.org --- Using TLA+ to Fix a Very Difficult glibc Bug - Malte Skarupke - C++Now 2025 --- The glibc condition variable (meaning std::condition_variable) was subtly broken for…

Using TLA+ to Fix a Very Difficult glibc Bug www.youtube.com/watch?v=Brgf...

GNU and the AI reimplementations antirez.com/news/162

Parma for Mortals A plain-language walkthrough of how Parma protects your containers from the people running them.

Parma for Mortals

Building a Database on S3 Hold your horses, though. I'm not unveiling a new S3-native database. This paper is from 2008. Many of its protocols feel clunky today. Yet...

Building a Database on S3 muratbuffalo.blogspot.com/2026/03/buil...

Rust zero-cost abstractions vs. SIMD A customer query was taking over 4× longer than it should have. The profiler pointed at Rust code we'd assumed was free. We followed the trail all the way down to assembly to find the true cost.

Rust zero-cost abstractions vs. SIMD

Beej's Guide to Network Programming: Using Internet Sockets beej.us/guide/bgnet/...

Inside the M4 Apple Neural Engine, Part 1: Reverse Engineering How we bypassed CoreML and talked directly to the hardware

Inside the M4 Apple Neural Engine, Part 1: Reverse Engineering

the mathematics of compression in database systems why compression is (almost) always worthwhile

the mathematics of compression in database systems

GenDB: The Next Generation of Query Processing — Synthesized, Not Engineered Content selection saved. Describe the issue below:

GenDB: The Next Generation of Query Processing — Synthesized, Not Engineered arxiv.org/html/2603.02...

Bespoke OLAP: Synthesizing Workload-Specific One-size-fits-one Database Engines Modern OLAP engines are designed to support arbitrary analytical workloads, but this generality incurs structural overhead, including runtime schema interpretation, indirection layers, and…

Bespoke OLAP: Synthesizing Workload-Specific One-size-fits-one Database Engines arxiv.org/abs/2603.02001

When AI Writes the World's Software, Who Verifies It? Leonardo de Moura — Creator of Lean and Z3

When AI Writes the World’s Software, Who Verifies It? leodemoura.github.io/blog/2026/02...

SFQ: Simple, Stateless, Stochastic Fairness

Turing Award Winner On Thinking Clearly, Paxos vs Raft, Working With Dijkstra | Leslie Lamport I interviewed Leslie Lamport, a Turing Award winner known for his contributions to distributed systems and the inventor of the Paxos algorithm. We walked through the major contributions of his career…

Turing Award Winner On Thinking Clearly, Paxos vs Raft, Working With Dijkstra

Modular: The Claude C Compiler: What It Reveals About the Future of Software Compilers occupy a special place in computer science. They're a canonical course in computer science education. Building one is a rite of passage. It forces you to confront how software actually…

Modular: The Claude C Compiler: What It Reveals About the Future of Software

BitFields API: Type-Safe Bit Packing for Lock-Free Data Structures Modern concurrent data structures increasingly rely on atomic operations to avoid the overhead of locking. A valuable but under-utilized technique for maximi...

BitFields API: Type-Safe Bit Packing for Lock-Free Data Structures rocksdb.org/blog/2025/12...

MariaDB innovation: binlog_storage_engine smalldatum.blogspot.com/2026/02/mari...

Zvec | A lightweight, lightning-fast, in-process vector database Meet Zvec

Zvec: A lightweight, lightning-fast, in-process vector database

Native Geospatial Types in Apache Parquet Native Geospatial Types in Apache Parquet

Native Geospatial Types in Apache Parquet

How to build a distributed queue in a single JSON file on object storage How to build a single global queue for distributed systems on object storage: Start with a single file on object storage, then add write batching, a stateless broker, and high-availability.

How to build a distributed queue in a single JSON file on object storage

Sourav Singh Rawat Systems engineer writing about databases, distributed systems, and low-level programming.

Redis/Valkey Replication Internals: The Architecture Behind Zero-Copy Command Propagation frostzt.com/blog/redis-v...

Vector search using only Parquet and DataFusion – Xiangpeng’s blog Just tune your existing stack

Vector search using only Parquet and DataFusion blog.xiangpeng.systems/posts/vector...

High-Throughput Graph Abstraction at Netflix: Part I By Oleksii Tkachuk, Kartik Sathyanarayanan, Rajiv Shringi

High-Throughput Graph Abstraction at Netflix: Part I netflixtechblog.medium.com/high-through...

PGO: Current state and challenges fosdem.org/2026/events/...

Go: Inside Reflection fosdem.org/2026/events/...

Delegating SQL Parsing to PostgreSQL fosdem.org/2026/events/...