First of all apologies for the missing two issues of this newsletter. I was away from Berlin and work for a couple of weeks, I assumed that there would have been time to put together reading and events… no chance, vacationing with a family is more demanding of a full time job (you cannot look for GIFs while preparing breackfast).
But enough about me and more about us. Next week on Thursday (August 27), with the friends from Gemma Analytics we are hosting the Data Berlin Summer Get Together. It’s a social event, the first drink is on us, just come for a chat and networking. In September we will be back with our usual meetups.
📖Interesting Readings
DuckDB turns into a client/server database in v2.0 — native Quack protocol, VARIANT as a first-class type, async I/O, a new SQL parser and a new storage format. – A Preview of DuckDB v2.0
A Rust rewrite of Postgres's query engine using batching, operator fusion and SIMD hits 300x on ClickBench, showing how the 1980s Volcano executor is the real bottleneck. – Rebuilding Postgres for 300x faster analytics
Shopify swapped Redis for MySQL for inventory reservations and found the relational database scaled better, a reminder that a cache isn't always the answer to a consistency problem. – We replaced Redis with MySQL for inventory reservations and it scaled
A forensic hunt through a silent SQLite data-corruption bug that took Tailscale 16 years to surface, showing how WAL checkpoint resets can quietly destroy committed data. – Tracking down the 16-year-old WAL-reset SQLite bug
A practical decision tree for choosing between full refresh, append, merge, delete+insert, time_interval, and SCD2 strategies based on source behavior rather than destination capabilities. – How to choose an incremental data loading strategy
An argument that the semantic layer has quietly become a control plane, and that vendors are rebuilding lock-in above the open bottom of the stack — the same trap open table formats fell into. – The Semantic Layer Is Not a Layer Anymore
Anthropic examines why agent swarms fail in shared environments — not because agents misbehave, but because their incentives diverge from human ones, pushing state isolation and protocol design to platform level. – Patterns and problems in emerging multiagent systems
🚀 Shipping Now
dbt Core v1.12 goes GA with an opt-in Rust parser that speeds up DAG compilation by 5-10x on large projects.
DataFusion Comet 1.0.0 accelerates Spark by translating physical plans to DataFusion native plans, with Spark 3.4-4.1 support.
Snowflake Power BI ingestion for Semantic View Autopilot hits GA.
dltHub for Teams makes agent-built pipelines safe to run with agentic alerts, team workspaces, and managed infrastructure.
Lightdash integrates with MotherDuck so the BI layer can query MotherDuck directly.
☕ Upcoming Meetups & Events
Aug 25 — Agents Gone Rogue 🤖
Aug 25 — Berlin Vibe Coding Club: Demo Nigh 🏛️
Aug 27 — Data Berlin Summer Get Together🐻
Sep 8 — Building AI Applications at Scale ft. ClickHouse, Grafana & dltHub 🏞️
Sep 14 — AWS UG Berlin Community Day Special ☁️
💼 This Week’s Job Picks
📊 Data Analyst
Product Data Analyst – FREE NOW → Apply here
Data Analyst – Planet → Apply here
Growth Analyst – Kayzen → Apply here
Data Analyst Recommerce – Zalando → Apply here
BI / Data Analyst – comrce group → Apply here
And more here.
📈 Analytics Engineer
Analytics Engineer, Payments Intelligence – Vinted → Apply here
Senior Analytics Engineer, Consumer OR Merchant – Wolt → Apply here
Senior Analytics Engineer - Shop Analytics – About You → Apply here
Senior Analytics Engineer – Alpaca → Apply here
Analytics Engineer (Decision Logic / Rules) – Enpal → Apply here
And more here.
⚙️ Data Engineer
Staff Data Engineer – Enpal → Apply here
Senior Data Engineer – celebrate company → Apply here
Senior Data Engineer - Lounge by Zalando – Zalando → Apply here
Data Engineer (6-month contract) – Fyul → Apply here
Data Engineer – Statista → Apply here
And more here.
🤖 AI / ML / LLM + Data Scientist
AI Infrastructure Engineer – Intercom → Apply here
Staff Forward Deployed Engineer - AI-Native Software – JetBrains → Apply here
Data Scientist, Buyer – Vinted → Apply here
Data Scientist – Kayzen → Apply here
Senior Decision Scientist, Customer Care Analytics – GetYourGuide → Apply here
And more here.
Ok, we are at the end of this issue. Sorry for letting you wait for the past two weeks (and half). In the meantime don’t forget that our website databerlin.net doesn’t sleep and continues to get daily updates (and more than 900 jobs).
And continue to send us you feedback, recently we discovered that we weren’t tracking correctly the R language, but now we do, thanks to one of our readers (you know who you are!).
That said, have a great reat of the week and I hope to see you on Thursday.
Bye,
— Data Berlin team