<!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;">OpenAI scaled PostgreSQL to handle millions of QPS for 800M users using one primary and ~50 read replicas, plus heavy use of query optimization </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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/AL2smj6RoRe1sJUx-X5O2NGC4i2B7_6e3DDRTFoJt34=442" 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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/z2bHyCghu0i0QWhzsKfahyDm2-p6Mj_SiPvAPi_M8f4=442" 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=edd873d8-fcc1-11f0-9a66-8d656233374f%26pt=campaign%26t=1769684868%26s=72194c5b1f842fab32b7a097f65cbff9c2b1c83e69336a44aa847d9a471e3963/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/UD4cDZiyZtjaQoFKC6RttyqIcfZN0J9pPDQvo_22PfY=442"><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></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">2026-01-29</span></strong></h1>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width:100%;" width="100%"><tbody></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%2Flinks.tldrnewsletter.com%2FJRmfFU/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/1_q1V5tAOS4i_Vw7HzCeRShjZi6yGUTlKWKo6gFPOHE=442">
<span>
<strong>Scaling PostgreSQL to power 800 million ChatGPT users (15 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
OpenAI scaled PostgreSQL to handle millions of QPS for 800 million users using one primary and ~50 read replicas, plus heavy use of query optimization, pooling, caching, and workload isolation. Write-heavy traffic is being moved to sharded systems like CosmosDB, with cascading replication and further architectural changes planned to extend scalability.
</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.linkedin.com%2Fblog%2Fengineering%2Fai%2Fcontextual-agent-playbooks-and-tools-how-linkedin-gave-ai-coding-agents-organizational-context%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/aUnIXaBGt5--KxDjpnCJ_6RvWGVQyqZCFhI5Y0uFM64=442">
<span>
<strong>Contextual Agent Playbooks and Tools: How LinkedIn Gave AI Coding Agents Organizational Context (8 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
LinkedIn's Contextual Agent Playbooks & Tools (CAPT) is a framework that equips AI coding agents with deep organizational context. Built on Model Context Protocol (MCP), CAPT provides agents with safe, standardized access to internal tools and composable "playbooks" (Jinja2-templated workflows that encode institutional knowledge as reusable step-by-step instructions), allowing dynamic chaining, reduced manual toil, and high developer adoption.
</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%2Fthe-ai-evolution-of-graph-search-at-netflix-d416ec5b1151%3Fgi=469a15a19643%26source=rss----2615bd06b42e---4%26utm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/bCNDSj6FRPGxDUYxZnzU6KLytOACX6L8B57nQR4h0nA=442">
<span>
<strong>The AI Evolution of Graph Search at Netflix: From Structured Queries to Natural Language (13 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Netflix shifted to advanced AI techniques for its graph-based search system, including dense vector embeddings for semantic similarity search, transformer-based models for query understanding and ranking, multimodal signals, and real-time graph traversals, to dramatically improve recall, relevance, personalization, and handling of complex/ambiguous queries.
</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%2Flinks.tldrnewsletter.com%2FdyPB4F/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/frTOWIxuVa46hNiVx8n74J1m9lfgiii5Jf38ULWdgBs=442">
<span>
<strong>Why We've Tried to Replace Data Analytics Developers Every Decade Since 1974 (15 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The data industry has repeatedly attempted to eliminate the need for data analytics/BI professionals through new tools. Yet, the fundamental complexity of turning raw business data into actionable insights requires human expertise in understanding context, modeling data accurately, handling transformations, ensuring governance, and iterating with business users.
</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%2Froundup.getdbt.com%2Fp%2Fapache-iceberg-and-the-catalog-layer%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/RLt2vYSbt2tonuwfx8ZxK3T2AfFD0zeMjC98YTcT7YQ=442">
<span>
<strong>Apache Iceberg and the Catalog Layer (53 minute podcast)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
The Iceberg catalog layer is the core metadata system that handles table discovery, conflict resolution, concurrency control, access governance, and cross-engine interoperability across Spark, Flink, Trino, Snowflake, and more. Modern REST-based catalogs like Apache Polaris enable pluggable, multi-cloud, vendor-neutral lakehouse architectures.
</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%2Fai-trends-reshaping-data-engineering-in-2026_602816%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/-YHw-1ssIYct-At6T1ni2N6Br1VL0ddmQxLln9_F8Ak=442">
<span>
<strong>AI Trends Reshaping Data Engineering in 2026 (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Key AI trends reshaping data engineering in 2026 include converging analytical and operational stacks to cut latency and costs, shifting to streaming-first real-time processing to tackle data quality issues, turning unstructured data into AI-ready assets with embeddings and vector search, evolving from prompt engineering to durable context engineering for institutional knowledge, and redesigning infrastructure for the massive recursive workloads of autonomous AI agents.
</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%2Fgithub.com%2Fsqlfluff%2Fsqlfluff%2Freleases%2Ftag%2F4.0.0%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/kog2V6cn4rALm7cN9HZzdYf4pPIWRb-B1IIXJqo_x7w=442">
<span>
<strong>SQLFluff 4.0 Released (8 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
SQLFluff 4.0 introduces an optional Rust-based parser and lexer for significant performance gains on large projects, with plans to make it the default in v5.0. The release also updates dbt support, fixes bugs, and expands SQL dialect coverage across engines like Postgres, BigQuery, DuckDB, and T-SQL.
</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.snowflake.com%2Fen%2Fblog%2Fopen-semantic-interchanges-specs-finalized%2F%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/6m9pvIJqtuD3ufm4akamgN69u1sBwb3I83Q21SIM4YI=442">
<span>
<strong>Open Semantic Interchange (OSI) Updates: Specification Now Live, New Working Group Members and More (3 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
OSI is a new open-source standard for sharing data semantics, so definitions like metrics and dimensions work consistently across tools and AI systems. The spec is live, the partner ecosystem is growing, and the project will move to foundation-led governance.
</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%2Ftobilg.com%2Fposts%2Fcustom-duckdb-wasm-builds-for-cloudflare-workers%2F%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/uvrKxARME0r0TZM1QUNSY6ZE-8GH25cspLClmqeWqao=442">
<span>
<strong>Custom DuckDB Wasm builds for Cloudflare Workers (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Ducklings enables full DuckDB SQL execution (JOINs, aggregation, and remote Parquet/CSV/JSON queries) directly within Cloudflare Workers, eliminating server/database dependencies and cold-start latency. Leveraging Asyncify, the package bridges DuckDB's synchronous HTTP file system with Workers' async fetch() API. The minimal WASM build (~9.6 MB gzipped) includes httpfs, Parquet, and JSON support, allowing edge workloads such as analytics, ETL, and dynamic SQL APIs with low-latency, global serverless deployment. Memory is limited to 128 MB, and only static extensions are supported.
</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%2Fdataengineeringcentral.substack.com%2Fp%2Flance-for-embeddings-vector-database%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/O8NGenu48QRcuGynNugxOoxpS2b68MFppteXU0_xUwo=442">
<span>
<strong>Lance for Embeddings (vector database) (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Lance offers a lightweight, file-based storage solution optimized for AI workloads, particularly embedding and vector storage, eliminating the complexity and operational overhead of traditional vector databases. The format integrates seamlessly with modern data tools like DuckDB and LangChain, supports similarity search natively, and requires only a basic pip install for setup.
</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.infoworld.com%2Farticle%2F4122336%2Fthe-private-cloud-returns-for-ai-workloads.html%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/qtx0k3SHIdKuAj9Q1WAdFUirmXb96jifTRtYWo1iDs8=442">
<span>
<strong>The private cloud returns for AI workloads (5 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Enterprises aggressively adopting public cloud for generative AI are facing unexpected costs, operational risks, and data proximity challenges due to spiky, GPU-intensive workloads and unpredictable scaling. AI-driven architectures expose inefficiencies—like persistent metered pricing and multi-service dependency chains—making private cloud deployments near operational centers financially and operationally preferable for inference and retrieval, while using public cloud for burst training.
</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.anthropic.com%2Fengineering%2Fdemystifying-evals-for-ai-agents%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/h8J-f8xd8MSGKlWZoKbEgioYg1Pd2dxU_qYt6zhVHsM=442">
<span>
<strong>Demystifying evals for AI agents (19 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Rigorous, automated evaluation frameworks are crucial for scalable, reliable AI agent development, transforming reactive debugging into proactive, metric-driven iteration. Effective agent evals combine code-, model-, and human-based grading across multidimensional tasks (e.g., coding, research, conversation, and GUI operations) using industry benchmarks and custom test suites. Quantitative methods like pass@k and pass^k provide nuanced performance metrics. Early and continuous investment in eval infrastructure accelerates delivery, prevents regressions, and creates actionable quality metrics. Open-source tools like Harbor, Promptfoo, and LangSmith lower adoption barriers.
</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.astronomer.io%2Fblog%2Fstate-of-airflow-2026%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/J60qwddSehwRnfzwu1RIsbQJz_eRlGw07tm-s4vsbg0=442">
<span>
<strong>State of Airflow 2026: The Orchestration Layer is Uniting Data, AI, and Enterprise Growth (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
26% of all Airflow users (and about half of Astronomer's) already upgraded to Airflow 3, less than a year post-release.
</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%2Fskey.uk%2Fpost%2Fkafka-dead-letter-queue-troubleshooting-guide%2F%3Futm_source=tldrdata/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/Iz0EkovtH8KPKYd4WN0FlOEbfcgyOfW1RrsWMSYFK3s=442">
<span>
<strong>Kafka Dead Letter Queue Triage: Debugging 25,000 Failed Messages (5 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
A practical guide to turning DLQs from silent data graves into early-warning systems for resilient streaming pipelines.
</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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/HkO82-4QBbs6VWzg5te-UCyzcXhOWITKrv2iDHnTw5w=442"><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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/AhmY1wi9z1v0eCZwSvy0uqP8DE6SYkKPmjlinFIlfw0=442" 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-d3e165357871/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/PJRLdS4wqhBniGYcVCMeIHDWz2DQfrlOUT7MAURoqEQ=442" 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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/XCkhE-yMYOeVNYydD5dQwGPMsA-ijN9UpHz94fxJWFA=442" 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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/oG5yTsKKTKj-gNg5vu3rDUIAlDWB1kY3IL39ZCGKj4o=442"><span>Joel Van Veluwen</span></a>, <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fjennytzurueyching%2F/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/4dyVBHg04dl_4yIOBUqq7nUa6AU_Bee5p_kLaZJnnz8=442"><span>Tzu-Ruey Ching</span></a> & <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fremi-turpaud%2F/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/dVRcEXaiRWcmT0KLZV2uYf-8GLg6Mr8ctmY8yo47lTc=442"><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/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/ohSW7b6VwIIXaLQwN_Yf1aG-K4nxXrtiI4dqggaSANw=442">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=edd873d8-fcc1-11f0-9a66-8d656233374f%26pt=campaign%26pv=4%26spa=1769684487%26t=1769684868%26s=2fa8354d5f56c6c8dc9537061478ea0e5c162f32dd180ee371b16b0a5e371598/1/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/qXZEl2BaOjBawjmHNLCpIDOeONkOFMi6nw3qyeALJZM=442">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/CI0/0100019c09701f99-f0da607f-7bce-45b9-8f9c-46abdea3f99d-000000/AmXvlRFwVoZD5UeBw3T9ICh_J9EELlDCSu8Gv4hPPmU=442" style="display: none; width: 1px; height: 1px;">
</body></html>