<!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;">DataFusion Comet speeds Spark reads on Iceberg by keeping Java planning while using Iceberg Rust for native Arrow execution. On a TPC-DS benchmark </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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/1q7WPqAcrZUjCGk7WUCEZ6_kC0CurSamoWMAxMjymy8=452" 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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/A1rXH24zVFDexDlVnNqDhHKEe590L0FLEAYQvPlAgDE=452" 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=b6fa887c-83f0-11f1-9a81-8befceb02188%26pt=campaign%26t=1784542423%26s=1fd865977b60f757b8ad99734ee497168bef900b21a69e886ba6e6de82e29654/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/VpK5XPofe73cPP5o2PyEclwYZDdoco7yW75GmDhYui8=452"><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%2Fusestring.ai%2F%3Futm_source=tldr%26utm_medium=newsletter%26utm_campaign=web-access-api-launch/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/IFZ9mG2bvd5K3WmplPjqr5OnNhYksQNv1tGvHhHePQk=452"><img src="https://images.tldr.tech/stringai.png" valign="middle" style="vertical-align: middle !important; height: 100%;" alt="String"></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">2026-07-20</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%2Fusestring.ai%2F%3Futm_source=tldr%26utm_medium=newsletter%26utm_campaign=web-access-api-launch/2/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/WIXgGDWBQJXpLhGqlJzN7FPDKYFQhG1c2I21awuKj8E=452">
<span>
<strong>Your agent is failing to fetch web data; the average failure rate for hard-target web fetches is 40%. Ours is 4%. (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The sites that block scrapers are exactly the ones data teams need, but most benchmarks ignore those. String's <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fgithub.com%2Fusestring%2Fweb-data-frontier-benchmark/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/9Ol-tVwl9xHIf21lZTVtqmmDUBQ-0gVMc30HP_Ld3M8=452" rel="noopener noreferrer nofollow" target="_blank"><span>open-source benchmark</span></a> looks at the hard cases.
<p></p>
<p><strong>The results</strong>: String's Web Access API hit<strong> 96% success on the hardest targets</strong>. Browserbase? 51%. Nimble? 59%. Oxylabs? 69%. Firecrawl? 72%.</p>
<p>Real Amazon prices, live Zillow comps, Indeed postings, if it's online, String will reach it, and we'll do it faster and cheaper than your current vendor.</p>
<p>Fetch, search, map, and browser endpoints. Use an API key or MCP.</p>
<p><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fportal.usestring.ai%2Fsign-up%3Futm_source=tldr%26utm_medium=newsletter%26utm_campaign=web-access-api-launch/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/9et1wUMKcfLZjI9rATTBX0YMRRgkLTIxi5eh9cf9iZU=452" rel="noopener noreferrer nofollow" target="_blank"><span>Try String for free here</span></a></p>
<p>Want to brainstorm your hardest use cases? <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fcal.com%2Frohit-shenoy-8pdwqu%2F15-min-meeting-all%3Futm_source=TLDR%26utm_medium=newsletter%26utm_campaign=web-access-launch/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/W1p2iJW7OacsnxDZYzdTTNNtroknzb4nWbkceGR7vFQ=452" rel="noopener noreferrer nofollow" target="_blank"><span>Book time with us and we'll solve them together</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%2Ficeberg.apache.org%2Fblog%2Faccelerating-iceberg-rust-development-with-datafusion-comet%2F%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/rXxWi5SFZldhh28jIrc3tZz5I8vIoTe9mxAt1cfba6k=452">
<span>
<strong>Accelerating Apache Spark Queries (and Iceberg Rust Development) with Apache DataFusion Comet (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
DataFusion Comet speeds Spark reads on Iceberg by keeping Java planning while using Iceberg Rust for native Arrow execution. On a 3 TB TPC-DS benchmark, it accelerated 102 of 103 queries and cut total runtime by about 40%, while its use of Iceberg Java's Spark test suites has helped drive more than 40 Iceberg Rust pull requests.
</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.infoq.com%2Fpresentations%2Fotel-slm-ai%2F%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/Pvdq9SzJfolB9gWrWWI94b-x_9q-e87iGXlcAkWauhE=452">
<span>
<strong>From OTEL to SLMs: Distilling Frontier Model Behaviour from Production Telemetry (45 minute video)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
OpenTelemetry traces can convert production AI-agent feedback into training labels for local 7B-13B coding models. Fine-tuned within hours, they can reach roughly 80-85% of frontier-model quality while lowering inference costs, preserving privacy, and scaling across the enterprise.
</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%2Fengineering.fb.com%2F2026%2F07%2F15%2Fai-research%2Fexploring-hierarchical-interest-representation-for-meta-ads-deep-funnel-optimization%2F%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/PO-q-JPceroG6I66sH0C8gioH_TQZemX4MWS2fD_Ags=452">
<span>
<strong>Exploring Hierarchical Interest Representation For Meta Ads Deep Funnel Optimization (8 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Meta compresses its graph of billions of users and advertiser entities into clusters of latent interests, enriches them with LLM-processed ad content, and embeds everything in one shared space where any user or entity can be scored against any other. Learning from broad graph structure instead of only rare conversion events gives the deep funnel a usable signal, and the embeddings feed Meta's ad ranking models like GEM and Andromeda.
</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%2Fwww.linkedin.com%2Fevents%2F7482821146388623361%2F%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/kWW1fN-4daUecolGiunY9rwap4jMgnXLoGX0dtojpgs=452">
<span>
<strong>Data Modeling isn't Dead, You Just Stopped Doing It with Joe Reis (65 minute video)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Data modeling is still essential, especially in the AI era, because clear structures, standards, and context help teams avoid bad data, wasted work, and constant firefighting. Modern data professionals need a flexible “mixed model” approach that uses the right modeling style for the problem rather than forcing everything into one method.
</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%2Fventurebeat.com%2Fdata%2Fagents-think-in-milliseconds-legacy-infrastructure-doesnt-linkedin-walmart-and-zendesk-shared-how-they-closed-the-gap-at-vb-transform-2026%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/FuCTIGMC0OQr44mnLtM379No_erav2LxEGdXJ4UBbJ4=452">
<span>
<strong>Agents think in milliseconds, legacy infrastructure doesn't. LinkedIn, Walmart, and Zendesk shared how they closed the gap at VB Transform 2026 (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Enterprise agents are constrained more by legacy infrastructure than model quality. LinkedIn pre-provisioned containers and shifted 80% of orchestration to deterministic code, Walmart governed duplicate internal agents, and Zendesk strengthened data pipelines for its 20B conversations. The lesson is to invest in evals, own the agent harness, and keep workloads portable across 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%2Fblog.opendataproducts.org%2Fai-agents-need-data-product-context-not-more-rag-99fa4690f4c8%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/GscLTLR4oSVyEeU6o4B-NVGD0xaAqgD4GLIhLnmY_CU=452">
<span>
<strong>AI Agents Need Data Product Context Not More RAG (11 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Enterprise AI agents need governed context, not just vector search: retrieval can find relevant passages, but it does not preserve source boundaries, business meaning, ownership, or usage restrictions. The proposed approach compiles one governed data-product model into multiple representations: graph, Markdown wiki, portable OKF bundle, and compact sidecars such as TOON/GCF. AI readiness depends on connected, trusted context and repeatable operations rather than catalog presence alone.
</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%2Fcube.dev%2F%3Futm_source=tldrdata%26utm_medium=newsletter%26utm_campaign=cube-general-promotion%26utm_content=v1-workflow/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/4FRe2Qvvg5XDKRrAsznCxOKvhv8TarVBJyXYi5Tpg8k=452">
<span>
<strong>😥 CEO waiting for data? Cube agents deliver so your team can focus on thinking (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Asking a data question is still a ticket, a queue, and a week of bogged down data teams. <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fcube.dev%2F%3Futm_source=tldrdata%26utm_medium=newsletter%26utm_campaign=cube-general-promotion%26utm_content=v1-workflow/2/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/22BkqQ-WXy03IvFcd5IlgSfGZLyJJhYp_3VCp7Lu0Mc=452" rel="noopener noreferrer nofollow" target="_blank"><span>Cube</span></a> hands that workflow to AI agents who model, explore, and build reports - working from one governed set of definitions and single source of truth. Brex, Wix, and Patagonia run on Cube. <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fcube.dev%2F%3Futm_source=tldrdata%26utm_medium=newsletter%26utm_campaign=cube-general-promotion%26utm_content=v1-workflow/3/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/2Vuqq7tsviwzeAohQARx9kAmwN-iqD5AxXwMTZig5zE=452" rel="noopener noreferrer nofollow" target="_blank"><span>Get started for free</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%2Fwww.databricks.com%2Fblog%2Fintroducing-apache-spark-42%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/oih-CC96LQ1XihxizCOIFB-XTODgIJlB4EVx_yZv0ZU=452">
<span>
<strong>Introducing Apache Spark 4.2 (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Apache Spark 4.2 adds metric views for defining business metrics, SQL vector similarity search, native geospatial types, and change data capture queries through a new CHANGES clause. Spark Connect and default Arrow optimized Python execution make the engine easier to reach. The release bundles over 1,900 commits from more than 260 contributors.
</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%2Fmicrosoft%2FOntology-Playground%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/2T59UrESeh8rGUxjYx4OQYsmlOpGQVfq_9mwKOMbUew=452">
<span>
<strong>Ontology Playground (GitHub Repo)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Microsoft's Ontology Playground is a free, open-source tool for visually exploring, building, importing and sharing ontologies for Microsoft Fabric IQ.
</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.crunchydata.com%2Fblog%2Fpostgres-19-compression-from-pglz-to-lz4%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/zt-W07jO8lFr8lTQXxq5vETE6PMp3gTKZKh2Bw8O3HU=452">
<span>
<strong>Postgres 19 Compression: from pglz to LZ4 (8 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Postgres 19 plans to switch default TOAST compression from pglz to LZ4, while keeping the same unified compression framework across heap storage, TOAST, and B-tree indexes. Variable-length types like TEXT, VARCHAR, BYTEA, and JSONB are compressed automatically via varlena metadata, with rows targeting ~2 KB before spilling out-of-line to TOAST. LZ4 is much faster than pglz in tests, often compresses better, and still fails fast on incompressible 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%2Farxiv.org%2Fhtml%2F2606.29823v1%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/a09LA0FCQeDEYoNKbmsuz8ENMZ9xQB7-ijhQIcR1RCQ=452">
<span>
<strong>Experience Graphs: The Data Foundation for Self-Improving Agents (28 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Trellis stores agent artifacts, rewards, search data, and causal history as persistent experience graphs accessible through SQL, graph, vector, and temporal queries. Meta found that reusing past experience matched performance about 10x faster and reduced successful-solution token costs by 52%, though excessive memory reuse reduced exploration.
</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%2Fnetflixtechblog.com%2Fin-house-llm-serving-at-netflix-a5a8e799ea2c%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/9_a-F1skOec5yq4WIXhvFVegRlT1usnNWVKlX7suyIc=452">
<span>
<strong>In-House LLM Serving at Netflix (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Netflix runs LLM serving end-to-end in-house, using a unified JVM-based platform with gRPC plus an OpenAI-compatible HTTP API, rather than relying on hosted APIs. The stack standardizes deployment through Triton-backed Model Scoring Service, using vLLM as the paved-path engine after re-benchmarking. It also adds custom packaging, zero-downtime rollouts, and patched guided decoding to prevent malformed JSON output.
</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%2FMROXyR/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/l9YBzt006xf_Xtn976LgwaEXCv27Um2R4zl6mHsJ-KQ=452">
<span>
<strong>From Weeks to a Day: How We Made LLM Evaluation Fast Enough to Iterate on (10 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Airbnb cut LLM evaluation cycles from weeks to under a day by caching generated references and judge scores, since over half of model outputs across candidates were identical strings and judge drift of about 1% per run. Small LoRA adapters with rank under 50 train in under an hour on a single GPU for same day fixes, and a final validation stage runs sampled traffic to catch failures.
</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%2Ftechcrunch.com%2F2026%2F07%2F17%2Fdatabricks-hits-188b-valuation-extending-its-run-as-ais-favorite-second-act%2F%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/eXFQFWYydTcaY2Dlgl-iH0c3OvLNQuMi5hGnFFfAQlc=452">
<span>
<strong>Databricks hits $188B valuation, extending its run as AI's favorite second act (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Databricks is reportedly raising $3 billion at a $188 billion valuation after successfully repositioning itself from a data platform into a major enterprise AI provider.
</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.starburst.io%2Fblog%2Fdremios-exit-is-the-clearest-sign-yet-that-lakehouse-only-wont-survive-ai%2F%3Futm_source=tldrdata/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/p3gksgeb8ee_YlJS9NQswElgHOJXLYQt_ile47OOCEU=452">
<span>
<strong>Dremio's Exit Is the Clearest Sign Yet That Lakehouse-Only Won't Survive AI (3 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Dremio's sale to SAP is presented as evidence that data platforms built around a single lakehouse will struggle with AI's demands for broad reach, freshness and unpredictable querying.
</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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/YqkM1qP5HxBMW61fjTcSjjptsKpLSq2LqwEcPWJ0WKA=452"><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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/4dDMaVfxe1y_a7TnWgWFj3Jdo_3LsQC3zaliVo7T0c0=452" 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-d3e165357871/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/CpKpvKxTKgwwcxK6_phiAPO5TnvfQn37z5n_1X0o0m8=452" 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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/AluqNKAk6OFoiOvgz54LFJIowHhTOEFKKl9WD3JRTWc=452" 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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/DReqlA-O-QMUofA3jc2MhT8meYjF7cqpGZsSIvsvcvk=452"><span>Joel Van Veluwen</span></a>, <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fjennytzurueyching%2F/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/dfz8QA9oOfpODMl9E_BDcLa_GLIJoXJEIfOJPmR5jOo=452"><span>Tzu-Ruey Ching</span></a> & <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fremi-turpaud%2F/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/7inWPGFk47Ax2TFOBjZkEGkVHpXEpKX2HF7N6E8l57o=452"><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/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/BB0SHDIpD2pMVGoF9pf7NVp90HvCagEGER-tBIf-BL4=452">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=b6fa887c-83f0-11f1-9a81-8befceb02188%26pt=campaign%26pv=4%26spa=1784541790%26t=1784542423%26s=597bde7f73c7c190237612b280aea69e41dc8584d65579a2e265915342862339/1/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/MMfQqP2IWzDs66CLH6rF74J6pNJWg9G3LRFxWB9DJw4=452">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/CI0/0100019f7f0469f3-63f1839a-0338-4d72-a5a6-27067ac2741f-000000/Ece-UTSqeVB8xBEoemKwn_4_zM0wc65kGoydu7gJwXc=452" style="display: none; width: 1px; height: 1px;">
</body></html>