Heikki Linnakangas
Member Of Technical Staff at Databricks
Antarctica
Summary
🤩
RockstarHeikki Linnakangas is a seasoned backend engineer with 18 years of experience specializing in PostgreSQL internals, database systems, and performance optimization. A co-founder of Neon, he worked on serverless Postgres features like pageserver refactors, logical replication improvements, and storage/compute separation before joining Databricks as a Member of Technical Staff. Heikki is an active open-source contributor on prominent projects such as pgvector (vector similarity for Postgres) and the PostgreSQL JDBC driver, where his work focused on HNSW indexing performance, bug fixes, and hardening against resource and concurrency issues. Comfortable in low-level C and system design, he repeatedly improves robustness and performance through careful refactors and addressing edge-case bugs like integer overflows. Based in Antarctica, he brings a pragmatic, detail-oriented approach and a knack for squeezing more performance and reliability out of critical database components.
18 years of coding experience
4 years of employment as a software developer
Github Skills (29)
filesystem10
postgresql10
back-end-development10
storage10
databases10
approximate-nearest-neighbor-search10
c1110
software-quality10
jdbc10
java10
c1710
javas10
performance-optimization10
rust10
relational-databases10
Programming languages (12)
Github contributions (5)
neondatabase/neon
Mar 2021 - Jan 2023
Neon: Serverless Postgres. We separated storage and compute to offer autoscaling, code-like database branching, and scale to zero.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2292 reviews, 904 commits, 1177 PRs in 1 year 10 months
Contributions summary:Heikki's commits primarily involve backend development tasks, focusing on the core components of a serverless PostgreSQL database. They made significant contributions to the pageserver, restoring and managing data within the database system, including a refactor of the existing storage model. Additionally, the user addressed issues related to the management and querying of data within the database, including improving the processing of logical replication. Furthermore, the user enhanced the robustness of the pageserver by incorporating and expanding several error handling and performance improvements, ensuring greater reliability.
neonrustserverlessautoscalingoffer
pgvector/pgvector
Sep 2023 - Oct 2024
Open-source vector similarity search for Postgres
Role in this project:Back-end Developer
Back-end DeveloperContributions:6 reviews, 22 PRs, 82 comments in 1 year 1 month
Contributions summary:Heikki primarily focused on optimizing and improving the performance of the pgvector extension for Postgres. Their contributions include enhancing HNSW index build performance by optimizing candidate selection, modifying the input processing for vector data, and removing unused code, such as the HnswSpool structure. Furthermore, the user addressed several issues, including integer overflows and compiler warnings. They also refactored code by removing unnecessary calls and including headers directly where needed.
approximate-nearest-neighbor-searchvectorssimilarity-searchmachine-learningsimilarity
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.