<!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;">The code mode pattern improves MCP tool usage by having the LLM write and execute a script that composes multiple tools in a sandbox </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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/PRXGmNDN_PfXtrkaW_O2LlfdBBhAyS8tlVdtyWZ6yac=447" 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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/ZeNxSk3aFecc3ykpQSytXdNHM8tLynNgqx6qHsuFDFI=447" 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=1f903dd0-1869-11f1-aaed-378b887271b8%26pt=campaign%26t=1772708868%26s=ba3a95b4965dde360e184cfc0abbe631710c58d806bdb0c3edc339acf6011233/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/Akqge71FbdSlXqNR1TyppVeNb4JDOkmrKbBW5DFgMRw=447"><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%2Fwww.starburst.io%2Finfo%2Fai-replaces-bi%2F%3Futm_source=tldr%26utm_medium=paid-email%26utm_campaign=aida/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/vk9kQDGbpYCNdGkrijiF4YoyLr9SCCluUaZmz5Utdno=447"><img src="https://images.tldr.tech/starburst.png" valign="middle" style="vertical-align: middle !important; height: 100%;" alt="Starburst"></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">2026-03-05</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%2Fwww.starburst.io%2Finfo%2Fai-replaces-bi%2F%3Futm_source=tldr%26utm_medium=paid-email%26utm_campaign=aida/2/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/FAZPJlnJrvPKs_I4i33ky5nTqRnK0aKtszW6ZWK-U-c=447">
<span>
<strong>The way people ask questions has changed. Can your analytics stack give answers? (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Dashboards were built for predefined questions and scheduled updates. But when every new question becomes a ticket, analytics slows the business.<p></p><p>With <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.starburst.io%2Finfo%2Fai-replaces-bi%2F%3Futm_source=tldr%26utm_medium=paid-email%26utm_campaign=aida/3/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/Zm4OOinzwoWudSTnsquJDY0i4LzDrgoDw0FLRx4kDoU=447" rel="noopener noreferrer nofollow" target="_blank"><span>AIDA, Starburst's AI data assistant</span></a>, everyone from analysts to CEOs can explore enterprise data in plain language, run complex analyses, and apply their own business rules in real time. </p>
<p>✅ No SQL</p>
<p>✅ No tickets</p>
<p>✅ <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.starburst.io%2Finfo%2Fai-replaces-bi%2F%3Futm_source=tldr%26utm_medium=paid-email%26utm_campaign=aida/4/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/lZIFDnahF2sW2v6vg0-XqDopXLc_z0Iic9qr-XMF0Cc=447" rel="noopener noreferrer nofollow" target="_blank"><span>No BI backlog</span></a></p>
<p>Ask a question, get an answer. Skip the dashboard. Keep going.</p>
<p><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.starburst.io%2Finfo%2Fai-replaces-bi%2F%3Futm_source=tldr%26utm_medium=paid-email%26utm_campaign=aida/5/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/9ShdTQo9WmOVUypyFvY4KkWRKuFzDmmEYUdfF4-ivCg=447" rel="noopener noreferrer nofollow" target="_blank"><span>Get a demo →</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%2Fgithub.blog%2Fengineering%2Farchitecture-optimization%2Fhow-we-rebuilt-the-search-architecture-for-high-availability-in-github-enterprise-server%2F%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/GYp0m0rRytOQLRxFzKxZ3Usy9Kiaq0jq1HzZ0sGhtto=447">
<span>
<strong>How we rebuilt the search architecture for high availability in GitHub Enterprise Server (5 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
GitHub rebuilt the search architecture using Elasticsearch's Cross Cluster Replication (CCR) to run independent single-node clusters per instance (primary and replicas), enabling durable persistence, asynchronous replication triggered after Lucene segments are created, custom workflows for setup and failover management, zero-downtime migrations, and automatic replica promotion for failover.
</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%2Fnetflixtechblog.com%2Foptimizing-recommendation-systems-with-jdks-vector-api-30d2830401ec%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/_LK4pNrFDURAUY1J1ddSGQlVbLmSUOQoPLN4Q5lfVqk=447">
<span>
<strong>Optimizing Recommendation Systems with JDK's Vector API (9 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Netflix reduced CPU utilization for its Ranker service's serendipity scoring feature from 7.5% to ~1% per node by re-architecting its scoring logic. Key optimizations included transitioning from O(M×N) scalar dot products to batched, cache-friendly matrix multiplies with flat buffers, leveraging the JDK Vector API for SIMD performance gains in pure Java, and eliminating unnecessary allocations. These changes yielded a 7% CPU drop, 12% latency reduction, and 10% improvement in CPU/RPS.
</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%2Fzero-waste-agentic-rag-designing-caching-architectures-to-minimize-latency-and-llm-costs-at-scale%2F%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/67V_9L4FMVj7nsslUfLh6iuDWte8uhkj_Q1_4UkKWfA=447">
<span>
<strong>Zero-Waste Agentic RAG: Designing Caching Architectures to Minimize Latency and LLM Costs at Scale (19 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
A validation-aware, two-tier caching strategy for production-grade RAG systems reduces LLM token costs by over 30% and slashes response times from ~36 seconds to milliseconds for semantically similar queries. Combining semantic caching (embedding-based, ~95% similarity) and retrieval caching (context/topic-level, >70%), the architecture addresses redundancy, data staleness, and cache invalidation via timestamp checks, SHA-256 fingerprinting, and predicate caching.
</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%2Fthenewaiorder.substack.com%2Fp%2Fsql-is-solved-heres-where-chat-bi%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/ZuV3F4_Q2HWoWEAD-kiiLu7-Y1qGZWT0ZFBx7rkprks=447">
<span>
<strong>SQL Is Solved. Here's Where Chat-BI Still Breaks (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Empirical testing of agentic chat-BI systems using BIRD and DABStep benchmarks revealed high SQL generation accuracy (over 70% correct on BIRD) but exposed critical failure nodes: ambiguous metric definitions, out-of-scope questions, and common-sense gaps. Context and rule files (e.g., RULES.md) help but induce compounding errors and overfitting as complexity grows. Iterative human-in-the-loop evaluation, structured error classification, deterministic metric definitions, and reproducible CI testing are essential for 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%2Fjoereis.substack.com%2Fp%2Fthe-reckoning-is-already-here%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/vkWTRSgNitk_au_24D-Oa3nd_l2Zjote-owlsHGRplc=447">
<span>
<strong>The Reckoning Is Already Here (3 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
AI tools are already replacing a lot of routine data engineering and analytics work right now (not in the future), so prioritize deep business understanding, irreplaceable domain expertise, strong community ties, and staying ahead by mastering the newest AI models.
</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%2Fseattledataguy.substack.com%2Fp%2Flayer-by-layer-we-built-data-systems%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/boAbVEO68IjSt72YIQtZy7OR-d16MwQh0dsS1Mr_Hic=447">
<span>
<strong>Layer by Layer, We Built Data Systems No One Understands (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The modern data stack has evolved into incomprehensible "fractal" complexity through endless layering of tools, driven by promises of "ease" that enable rapid prototyping but foster departmental silos, decision avoidance, unchecked AI/LLM code generation, business logic over-modeling, and disconnection from real business value.
</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%2Fmoderndata101.substack.com%2Fp%2Fdata-products-as-ai-agents%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/CCKzugWWhc8xFSbIRtomltii7MfO1XXy95dHjxyqUwk=447">
<span>
<strong>How Long Until We Call AI Agents Data Products (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
AI agents in production must be managed as full-fledged data products, requiring rigorous observability, security, and iterative product analytics beyond standard logging. Treating agent interactions as actionable feedback loops drives roadmap decisions, while layered security and conversational discoverability are essential for user trust and adoption.
</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%2Fwww.jlowin.dev%2Fblog%2Ffastmcp-3-1-code-mode%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/E8lio1S0i6VaGK5iHY6uStVNGy2dPpT3PEpTi9GD4Bc=447">
<span>
<strong>Stop Calling Tools, Start Writing Code (Mode) (8 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The code mode pattern improves MCP tool usage by having the LLM write and execute a script that composes multiple tools in a sandbox, instead of calling tools sequentially. This reduces context window bloat and round-trip overhead, making large tool catalogs far more scalable and efficient for LLMs to use.
</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.alibabacloud.com%2Fblog%2Fpostgresql-blink-tree-implementation_602913%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/AlIEgYNO2yeWyPpdDr6MRRlRLliJHDptzlN5aYmC950=447">
<span>
<strong>PostgreSQL Blink-tree Implementation (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
PostgreSQL implements a high-concurrency version of B-tree indexes called Blink-Tree, adding a simple "link" pointer between sibling nodes and a "high-key" boundary marker in each node. This lets searches move quickly to the right sibling if needed without holding locks across multiple levels (no lock-coupling during reads), while structure changes like page splits use brief bottom-up lock-coupling on just a few nodes at a time, reducing lock contention dramatically.
</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%2Fgithub.com%2Fvladich%2Fpg_jitter%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/i2T71JjevstvAqGxAn7WKPcExjqbCvsiXNaPsXeNCYA=447">
<span>
<strong>PgJitter (GitHub Repo)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
PgJitter is a lightweight PostgreSQL extension that replaces the default LLVM JIT compiler with faster alternatives (sljit, AsmJIT, and MIR), enabling native code generation in microseconds instead of milliseconds. This dramatically reduces compilation overhead and makes JIT practical for a wider range of queries, especially OLTP workloads.
</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%2Fdatasciencexai.substack.com%2Fp%2Fai-evals-in-the-real-world-human%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/6WCKKt6B7BdsCmuYOQrGUZ-DMIta-F_1FH3sW3qGeNI=447">
<span>
<strong>AI Evals in the Real World: Human Judging, LLM Judges, and the Gaps Between (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Regular AI test scores don't work well for customer-service bots that need to keep conversations going, understand hidden intent, and actually get users to share contact info. The team built a better scoring system that mixes human taste-testing for tricky parts with LLM-as-judge auto-scoring for scale, plus human spot checks on bad cases.
</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%2Fsimonwillison.net%2F2026%2FMar%2F4%2Fqwen%2F%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/3hJdxfAD5ZewE8lIhakL2JPs6lLNz-AGYQT6pUp679I=447">
<span>
<strong>Something is afoot in the land of Qwen (5 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The Qwen 3.5 open-weight model family from Alibaba is gaining attention for delivering strong performance across a wide range of model sizes, including very small models that run locally while still supporting reasoning and multimodal tasks. However, the project's future is uncertain after the sudden resignation of its lead researcher and several core team members following an internal Alibaba reorganization.
</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.confluent.io%2Fblog%2Fkafka-queue-semantics-share-consumer-ga%2F%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/Sz2J7BEDIY_g3lZjK98CRFh8gjT94sjrX_1oTuTT0YA=447">
<span>
<strong>Queues for Apache Kafka Is Here: Your Guide to Getting Started in Confluent (11 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Queues for Kafka is now generally available on Confluent Cloud and will be released shortly on Confluent Platform, introducing queue semantics and elastic consumer scaling natively to Kafka via KIP-932.
</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%2Fdelta.io%2Fblog%2F2026-02-02-delta-catalog-managed-tables%2F%3Futm_source=tldrdata/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/USPMcgGVSOkuoaCnjS5VwtF6ZttlDpmdwQ9UqqOJqHI=447">
<span>
<strong>The next evolution of Delta - Catalog-Managed Tables (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Delta Lake 4.1.0 introduces catalog-managed tables, shifting control from filesystem paths to a central catalog for metadata, governance, and commits, improving discovery and cross-engine interoperability.
</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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/22rdgbfgVplEhVOKVDSVR3hz3qW6iDsgw11RKhjro_I=447"><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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/qtDhR_aJZ0_uZtzF90FKO6Xu-8QWsf46kbgllDYuKBg=447" 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-d3e165357871/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/WBOUdrFWmEC9QV9rx9lIgE3Gmsr7i4jyDLXO_Nj3r08=447" 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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/cY7yPRQeNdjEzqtknygeHWz2hCUQbdQruh1e8AnBLdQ=447" 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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/NWcd2YcU7eF0RWIHawQYeQ2BBnkVZ2kIfnbMQKZKTq0=447"><span>Joel Van Veluwen</span></a>, <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fjennytzurueyching%2F/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/HrgQZFRdLQNjzWJR1qQuNQkfEJPYOFitNrABPshw5DY=447"><span>Tzu-Ruey Ching</span></a> & <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fremi-turpaud%2F/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/lkkxkG87k-ZilhOhyVaCYnF1fUwf9HGJKBaWvIUOf5g=447"><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/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/LLDf2uS8ZX1uVJrLil1aHERu0iY66TF45RqmHmu6tD8=447">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=1f903dd0-1869-11f1-aaed-378b887271b8%26pt=campaign%26pv=4%26spa=1772708472%26t=1772708868%26s=d9bba85920704c1ee7aea672a5de7962a246dd5829f9260de120810b0d9f893e/1/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/oIH0NkXd1G1zoB6ZQWRLxeIu1m2V2JWkl_stAaVrK2Y=447">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/CI0/0100019cbdaeafe2-33bd4e05-0ada-4b93-913f-93ff9c831a31-000000/dACQP7uojFv1VyP-kvxq3fTi5amQqArsPRUeftSLD04=447" style="display: none; width: 1px; height: 1px;">
</body></html>