<!DOCTYPE html><html lang="en"><head><meta http-equiv="Content-Type" content="text/html charset=UTF-8"><meta charset="UTF-8"><meta name="viewport" content="width=device-width"><meta name="x-apple-disable-message-reformatting"><title>TLDR Data</title><meta name="color-scheme" content="light dark"><meta name="supported-color-schemes" content="light dark"><style type="text/css">
:root {
color-scheme: light dark; supported-color-schemes: light dark;
}
*,
*:after,
*:before {
-webkit-box-sizing: border-box; -moz-box-sizing: border-box; box-sizing: border-box;
}
* {
-ms-text-size-adjust: 100%; -webkit-text-size-adjust: 100%;
}
html,
body,
.document {
width: 100% !important; height: 100% !important; margin: 0; padding: 0;
}
body {
-webkit-font-smoothing: antialiased; -moz-osx-font-smoothing: grayscale; text-rendering: optimizeLegibility;
}
div[style*="margin: 16px 0"] {
margin: 0 !important;
}
table,
td {
mso-table-lspace: 0pt; mso-table-rspace: 0pt;
}
table {
border-spacing: 0; border-collapse: collapse; table-layout: fixed; margin: 0 auto;
}
img {
-ms-interpolation-mode: bicubic; max-width: 100%; border: 0;
}
*[x-apple-data-detectors] {
color: inherit !important; text-decoration: none !important;
}
.x-gmail-data-detectors,
.x-gmail-data-detectors *,
.aBn {
border-bottom: 0 !important; cursor: default !important;
}
.btn {
-webkit-transition: all 200ms ease; transition: all 200ms ease;
}
.btn:hover {
background-color: #f67575; border-color: #f67575;
}
* {
font-family: Arial, Helvetica, sans-serif; font-size: 18px;
}
@media screen and (max-width: 600px) {
.container {
width: 100%; margin: auto;
}
.stack {
display: block!important; width: 100%!important; max-width: 100%!important;
}
.btn {
display: block; width: 100%; text-align: center;
}
}
body,
p,
td,
tr,
.body,
table,
h1,
h2,
h3,
h4,
h5,
h6,
div,
span {
background-color: #FEFEFE !important; color: #010101 !important;
}
@media (prefers-color-scheme: dark) {
body,
p,
td,
tr,
.body,
table,
h1,
h2,
h3,
h4,
h5,
h6,
div,
span {
background-color: #27292D !important; color: #FEFEFE !important;
}
}
a {
color: inherit !important; text-decoration: underline !important;
}
</style><!--[if mso | ie]>
<style type="text/css">
a {
background-color: #FEFEFE !important; color: #010101 !important;
}
@media (prefers-color-scheme: dark) {
a {
background-color: #27292D !important; color: #FEFEFE !important;
}
}
</style>
<![endif]--></head><body class="">
<div style="display: none; max-height: 0px; overflow: hidden;">Dashboards provide visibility but not real analysis. “Self-service” plateaued because dashboards cannot reason or investigate </div>
<div style="display: none; max-height: 0px; overflow: hidden;">
<br>
</div>
<table align="center" class="document"><tbody><tr><td valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" class="container" width="600"><tbody><tr class="inner-body"><td>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr class="header"><td bgcolor="" class="container">
<table width="100%"><tbody><tr><td class="container">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" style="margin-top: 0px;" width="100%"><tbody><tr><td style="padding: 0px;">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div style="text-align: center;">
<span style="margin-right: 0px;"><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ftldr.tech%2Fdata%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/cYJkVPvKghUqYUp05jdyh3ftPq_yIQ-1hcsFHyZtdEI=435" rel="noopener noreferrer" target="_blank"><span>Sign Up</span></a>
|<span style="margin-right: 2px; margin-left: 2px;"><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fadvertise.tldr.tech%3Futm_source=tldrdata%26utm_medium=newsletter%26utm_campaign=advertisetopnav/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/qFwMgiMHn1u8IcmIc9zij0WlEyEc1N71_xqC7Tz_vEY=435" rel="noopener noreferrer" target="_blank"><span>Advertise</span></a></span>|<span style="margin-left: 2px;"><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fa.tldrnewsletter.com%2Fweb-version%3Fep=1%26lc=1670a604-84b7-11f0-bcf5-55fc1d40139c%26p=65db08f0-d986-11f0-95ae-2734f7cbc04d%26pt=campaign%26t=1765796800%26s=ec24ad0c49a3dffb9c49e5e8fd97132db619094206a806c20b8eef9775383aa2/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/R6maOEBBHJsR2aKYaFTLv1pRxWCfsDVHqGiuj7tqInc=435"><span>View Online</span></a></span>
<br>
</span></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="text-align: center;"><span data-darkreader-inline-color="" style="--darkreader-inline-color:#3db3ff; color: rgb(51, 175, 255) !important; font-size: 30px;">T</span><span style="font-size: 30px;"><span data-darkreader-inline-color="" style="color: rgb(232, 192, 96) !important; --darkreader-inline-color:#e8c163; font-size:30px;">L</span><span data-darkreader-inline-color="" style="color: rgb(101, 195, 173) !important; --darkreader-inline-color:#6ec7b2; font-size:30px;">D</span></span><span data-darkreader-inline-color="" style="--darkreader-inline-color:#dd6e6e; color: rgb(220, 107, 107) !important; font-size: 30px;">R</span>
<br>
</td></tr></tbody></table>
<br>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr id="together-with"><td align="center" height="20" style="vertical-align:middle !important;" valign="middle" width="100%"><strong style="vertical-align:middle !important; height: 100%;">Together With </strong>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FFAR5RM/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/-h5U-ploj63bv3JRB6IFvv1O-VSs4WUoR-CT7rEtqHA=435"><img src="https://images.tldr.tech/databricks.png" valign="middle" style="vertical-align: middle !important; height: 100%;" alt="Databricks"></a></td></tr></tbody></table>
<table style="table-layout: fixed; width:100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;">
<div style="text-align: center;">
<h1><strong>TLDR Data <span id="date">2025-12-15</span></strong></h1>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width:100%;" width="100%"><tbody><tr id="sponsy-copy"><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FFAR5RM/2/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/Hn_Zl0AqVFkhyNheuHAo_-qyCAOUC9A5QC4_QwzOB3Y=435">
<span>
<strong>Agents that don't suck (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Most AI agents never reach production. <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FFAR5RM/3/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/dguZk241e8qPRWmH4t2t2Bc3TB9RT8PoV-FD7ea70Aw=435" rel="noopener noreferrer nofollow" target="_blank"><span>Agent Bricks by Databricks</span></a> fixes that by helping you build agents that actually work — accurate, reliable and grounded in your data.
<p></p>
<p>This product delivers practical quality, not hype: automatic evaluation, metrics tuned to your goals, and human feedback to keep improving accuracy. Generic benchmarks don't cut it. <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FFAR5RM/4/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/tQ6hPho50Q8Yk_vZLPhRTBCvmJeTfL_I5q3pQnlOvkE=435" rel="noopener noreferrer nofollow" target="_blank"><span>Agent Bricks</span></a> measures performance on the tasks that matter to your business.</p>
<p>The result: production agents you can trust — built fast, governed from day one and ready for real workloads.</p>
<p><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FFAR5RM/5/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/M9TnoDiD-fCzJyAQvHpgMbYFU3Z5wDCMjXLYxCQYrgk=435" rel="noopener noreferrer nofollow" target="_blank"><span>See why it's worth your time</span></a>
</p>
</span></span></div>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr>
<tr bgcolor=""><td class="container">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td style="padding: 0px;">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">📱</span></div></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Deep Dives</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flevelup.gitconnected.com%2Fhow-to-measure-similarity-between-sql-queries-using-embeddings-36a74b4beabf%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/cj1leLudCbJToZtVV7pQ9dv29_SIaK6eiYGCkvRrnVQ=435">
<span>
<strong>How to Measure Similarity Between SQL Queries Using Embeddings (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Transforming SQL queries into vector embeddings enables precise, quantifiable measurement of query similarity, improving visibility into user behaviors and data access patterns in data lakes. Using sentence transformers (eg. all-MiniLM-L6-v2) to create embeddings and a vector database backend, queries are efficiently clustered, compared via cosine similarity, and visualized using methods like t-SNE to reduce dimensionality, revealing patterns such as tight clusters and outliers. This approach reduces guesswork in SQL optimization and workload analysis. Code linked in the article.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fblog.vectorchord.ai%2Fhow-we-made-100m-vector-indexing-in-20-minutes-possible-on-postgresql%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/9qyVKfJh5nZyF72QCji_mBmTuDYk7YCbl3yMRW8egtk=435">
<span>
<strong>How We Made 100M Vector Indexing in 20 Minutes Possible on PostgreSQL (13 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
VectorChord shows that 100M-scale vector indexing on PostgreSQL is practical without GPUs, reducing build time to ~20 minutes and memory to ~12 GB versus ~40 hours and ~200 GB with pgvector. The gains come from hierarchical K-means with dimensionality reduction and deep Postgres contention and parallelism optimizations, enabling low-cost, billion-scale vector search with minimal recall loss.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ftowardsdatascience.com%2Fspectral-community-detection-in-clinical-knowledge-graphs%2F%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/aIxlQKMDuQqYrNOnsY-irLa3g7YXVJKSlqoTOkvKi7Y=435">
<span>
<strong>Spectral Community Detection in Clinical Knowledge Graphs (22 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Graph algorithms can be layered with spectral methods to reveal deep, clinically meaningful structural insights in patient populations that traditional clustering would miss. Given a graph dataset of patients, diseases, and relationships, the Leiden algorithm can identify distinct communities, spectral analysis adds intra-community signal to quantify cohesion, and the Fiedler vector cleanly splits “mixed” communities. This pattern turns KGs into measurable, comparable cohort-structure features for exploratory analytics and GraphRAG-style enrichment. The code is available in the linked GitHub repository.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">🚀</span></div>
</div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Opinions & Advice</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fhex.tech%2Fblog%2Fdashboards-were-never-the-destination%2F%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/eT-nOmizGGa5a43bOLa3MvU4ivkIjoBhYzxgpsfpNLc=435">
<span>
<strong>Dashboards Were Never the Destination (8 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Dashboards were a workaround for limited tools. They provide visibility but not real analysis. “Self-service” plateaued because dashboards cannot reason or investigate. Agentic analytics shifts data teams from building static artifacts to curating context and logic, enabling systems that deliver real answers directly on the warehouse.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.ssp.sh%2Fblog%2Fomakase-data-stack%2F%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/nlPbYTmfLKtIp_Eb5pEwKAyki7tMIcTTdw6Y615DZjM=435">
<span>
<strong>Opinionated Data Platforms vs. Open-Source: The Chef's Choice for Your Data Platform (13 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Market consolidation is accelerating a shift from open, composable data stacks to unified, opinionated data platforms that deliver immediate end-to-end analytics, automated orchestration, and robust DevOps integration out of the box. These platforms significantly reduce the need for manual infrastructure management, integration, and troubleshooting, focusing teams on delivering business value rather than tool-wrangling. Open standards such as Iceberg, Delta Lake, and Apache Arrow remain crucial, ensuring portability and mitigating vendor lock-in even within closed-source platforms. Deploy open-source stacks when maximum flexibility is required, but leverage opinionated platforms for operational simplicity, rapid time-to-value, and seamless scaling.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fthenewstack.io%2Fai-agents-create-a-new-dimension-for-database-scalability%2F%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/nEy2sN0OSd6qh6StVUMcUgsfLptV8CP9fJ71tt0chtk=435">
<span>
<strong>AI Agents Create a New Dimension for Database Scalability (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The rise of AI agentic systems introduces a new scalability axis for databases: the ability to spawn and manage trillions of isolated, ephemeral database instances with microsecond provisioning and strict data isolation. Traditional multitenancy is giving way to hyper-tenancy, demanding in-process databases like SQLite, Turso, and DuckDB that offer instant instantiation, encryption, and vector-native operations. This architecture shift enables agents to fulfill dynamic, privacy-preserving data needs at hyperscale, transforming database design and operational paradigms for data-intensive organizations.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">💻</span></div>
</div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Launches & Tools</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fgetbruin.com%2F%3Futm_source=tldr%26utm_medium=sp%26utm_campaign=founders_202511/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/W8cuabThBw3ff1Uot9JGYJtphZmQ2jpeYADfRBHVKao=435">
<span>
<strong>Become an AI data engineer in 10 minutes with Bruin: the AI data vibeline (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Vibe coding doesn't work for data? It does now: <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fgithub.com%2Fbruin-data%2Fbruin/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/eJUMPfqKgUM7W38CQTTuptYgM58lkaJdI3OL-Q9HrRQ=435" rel="noopener noreferrer nofollow" target="_blank"><span>Bruin</span></a> is a command-line data tool that lets you build SQL & Python pipelines with built-in quality checks, column-level lineage, and end-to-end observability. It's open source, built in Go, and combines all the best parts of dbt + Airflow (minus the aggravation). <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fgetbruin.com%2F%3Futm_source=tldr%26utm_medium=sp%26utm_campaign=founders_202511/2/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/CldQFEzqXvFTfrz6gmsaKvvmQ5NyBj9clp2XsOw6NG4=435" rel="noopener noreferrer nofollow" target="_blank"><span>Build your first data vibeline</span></a>
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fclickhouse.com%2Fblog%2Fintroducing-pg_clickhouse%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/Nu7_PnR47buRbd9wx4pVbj0L8HOXr5IE-V4FWvM9P44=435">
<span>
<strong>Introducing pg_clickhouse: A Postgres Extension for Querying ClickHouse (5 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
pg_clickhouse v0.1.0, a PostgreSQL extension with near-universal query pushdown, enables seamless execution of analytics queries on ClickHouse directly from PostgreSQL, dramatically simplifying migration of analytical workloads. Supporting advanced pushdown for aggregates, FILTER expressions, and SEMI-JOINs, the extension streamlines moving from Postgres to ClickHouse, achieving sub-second TPC-H query execution in 21 of 22 cases. This eliminates costly query rewrites and leverages ClickHouse's high-performance engine for SQL-based analytics.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fopensource.googleblog.com%2F2025%2F11%2Fannouncing-magika-10-now-faster-smarter.html%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/vmounKC3hXvnGHTp43F9jw3bL8dO9qxjV-NS3TzH9Xs=435">
<span>
<strong>Announcing Magika 1.0: Now Faster, Smarter, and Rebuilt in Rust (5 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Google's Magika 1.0, now rebuilt in Rust, delivers AI-driven file type detection with support for over 200 formats—doubling coverage, with substantial gains in distinguishing nuanced text, code, and data science files. Leveraging a 3TB+ training set and ONNX Runtime, Magika scans nearly 1,000 files/sec/core on modern CPUs while achieving ~99% precision and recall. Enhanced granularity and robust memory safety significantly improve pipeline reliability.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2F1eTb8N/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/2AaKF2E7ajEZhsn0G0vXeofGPbTAtCHuUQmZT3nbPoI=435">
<span>
<strong>What it Means to Get Your Data Ready for AI (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Agentic AI shifts data engineering from feeding rigid dashboards to enabling agents that turn user intent into actions. Five recommendations: favor business context over deep normalization; curate a small, high-quality “example store” for in-context learning; expose tool-ready APIs and semantically friendly formats; manage agent outputs as versioned first-class data; and connect observability to automated retraining. Net: build a flexible, auditable platform where agents can compose workflows.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">🎁</span></div></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><strong><h1>Miscellaneous</h1></strong></div>
</div>
</td></tr></tbody></table>
<table bgcolor="" style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Ffeed%2Fupdate%2Furn:li:activity:7405691431547195392%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/rWu3oYJs8dBYWhE_-f6iGLX_OUFELRBaXzAxUIQKTWI=435">
<span>
<strong>Why XGBoost and LightGBM Are Secretly Leaking Your Data (and How CatBoost Fixes it) (2 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
CatBoost avoids a common but underdiscussed source of leakage in XGBoost and LightGBM by using ordered target statistics for high-cardinality categorical features, preventing models from memorizing rare categories. It also auto-generates categorical feature interactions, reducing manual feature engineering while improving generalization.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.youtube.com%2Fwatch%3Fv=RnQKluxWB5Y%26t=7s%26utm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/flWchYW06gqRIa1qcTSmx1Wy8penNklBDL0wF54nTQQ=435">
<span>
<strong>DBOS: 2025 Year in Review with Mike Stonebraker and Andy Pavlo (1 hour video)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
LLMs perform well on public SQL benchmarks but fail on real enterprise warehouses, with accuracy near zero in private tests and only ~30% even with RAG or decomposition. As AI-generated code increasingly touches databases, the stack is shifting toward agentic, ACID-durable systems with holistic autotuning, while Postgres consolidation and native vector support are eroding the standalone vector-database moat.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">⚡</span></div></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Quick Links</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table bgcolor="" style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.youtube.com%2Fwatch%3Fv=wIOYAIgPdqw%26utm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/q8UJMt1OyKJFILMINBmHyRfyDQzSz4wX6gHzCOgZwgg=435">
<span>
<strong>Scaling DuckDB in the Cloud with MotherDuck CEO Jordan Tigani (65 minute podcast)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
MotherDuck is a cloud service for DuckDB that scales to zero, isolates tenants per user, and supports hybrid execution that can split queries across local and cloud data.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.percona.com%2Fblog%2Fsupport-for-crunchy-hardened-postgresql-ends-soon-dont-get-caught-off-guard%2F%3Futm_source=tldrdata/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/GHgbf1CzyATwex63p_F_isTQw6A7y4IByIXxF_X1M6s=435">
<span>
<strong>Support for Crunchy Hardened PostgreSQL Ends Soon (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Regulated sectors must plan migrations before April 2026 to avoid compliance and patching gaps.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td align="left" style="word-break: break-word; vertical-align: top; padding: 5px 10px;">
<p style="padding: 0; margin: 0; font-size: 22px; color: #000000; line-height: 1.6; font-weight: bold;">
Want to advertise in TLDR? 📰
</p>
<div class="text-block" style="margin-top: 10px;">
If your company is interested in reaching an audience of data engineering professionals and decision makers, you may want to <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fadvertise.tldr.tech%2F%3Futm_source=tldrdata%26utm_medium=newsletter%26utm_campaign=advertisecta/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/XpkCCmqMQ6Opry_L46990goHZDQBvdyk_Rx6JUVYUDs=435"><strong><span>advertise with us</span></strong></a>.
</div>
<br>
<!-- New "Want to work at TLDR?" section -->
<p style="padding: 0; margin: 0; font-size: 22px; color: #000000; line-height: 1.6; font-weight: bold;">
Want to work at TLDR? 💼
</p>
<div class="text-block" style="margin-top: 10px;">
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fjobs.ashbyhq.com%2Ftldr.tech/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/opGeq75vY9dckDjBQXpS91UVMW0kIOft8i2U4-zsYOo=435" rel="noopener noreferrer" style="color: #0000EE; text-decoration: underline;" target="_blank"><strong>Apply here</strong></a>,
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fjobs.ashbyhq.com%2Ftldr.tech%2Fc227b917-a6a4-40ce-8950-d3e165357871/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/mYd9_vkFDHT6Ct85LFu5HXzUE9Ed66tN34bEQRUhVCA=435" rel="noopener noreferrer" style="color: #0000EE; text-decoration: underline;" target="_blank"><strong>create your own role</strong></a> or send a friend's resume to <a href="mailto:jobs@tldr.tech" style="color: #0000EE; text-decoration: underline;">jobs@tldr.tech</a> and get $1k if we hire them! TLDR is one of <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Ffeed%2Fupdate%2Furn:li:activity:7401699691039830016%2F/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/sspHFFCKM4RjLxbpn40HekUUpE9G5bmMEkx07VcSCrg=435" rel="noopener noreferrer" style="color: #0000EE; text-decoration: underline;" target="_blank"><strong>Inc.'s Best Bootstrapped businesses</strong></a> of 2025.
</div>
<br>
<div class="text-block">
If you have any comments or feedback, just respond to this email!
<br>
<br> Thanks for reading,
<br>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fjoelvanveluwen%2F/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/sTfcmafoDsbtYxhuQxMKbOCbMvInHU8MCkdBCeXRuUk=435"><span>Joel Van Veluwen</span></a>, <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fjennytzurueyching%2F/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/qlkgEPsi7tsp72lXf3UBGao-ouRvuNDNrpx6BnJUbD4=435"><span>Tzu-Ruey Ching</span></a> & <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fremi-turpaud%2F/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/GA6SjgBDA9H1DomyFtOCr5ALVksuhnjdTpLbe7g6qmM=435"><span>Remi Turpaud</span></a>
<br>
<br>
</div>
<br>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block" id="testing-id">
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ftldr.tech%2Fdata%2Fmanage%3Femail=silk.theater.56%2540fwdnl.com/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/vg3MI_uzPAGFFuuldG4HtR1OKq4nDMkLYKPTrFUMxMM=435">Manage your subscriptions</a> to our other newsletters on tech, startups, and programming. Or if TLDR Data isn't for you, please <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fa.tldrnewsletter.com%2Funsubscribe%3Fep=1%26l=037ede50-92cc-11ee-b0f2-b761aa2217ad%26lc=1670a604-84b7-11f0-bcf5-55fc1d40139c%26p=65db08f0-d986-11f0-95ae-2734f7cbc04d%26pt=campaign%26pv=4%26spa=1765796472%26t=1765796800%26s=513c186e278281b08aa43a4ffb8b5e53d77aa30f3cd3ca14d2617979e04b7b08/1/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/r_IuoH0RzH8ZnFthAm5WbliRR6zGbFRj0s9bKzMqB0o=435">unsubscribe</a>.
<br>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
<img alt="" src="http://tracking.tldrnewsletter.com/CI0/0100019b21b0e712-21f3761c-c499-431a-a6c2-a7b553edaa51-000000/zK0uXUXcPB7ze9KMzPT_JRPTgyB2zHTa3BnUTon022w=435" style="display: none; width: 1px; height: 1px;">
</body></html>