<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:podcast="https://podcastindex.org/namespace/1.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
<title>Modern Data Engineering</title>
<description>Somebody asks for revenue by region for the last three years, and it turns out the answer can&apos;t come straight from the tills. Modern Data Engineering is about the work behind answers like that: copying data from where it is born to where it is useful without breaking, losing or leaking it. Shelley has a boss who wants every dashboard live and a neighbour, Ines, who is a data engineer at a grocery chain called Bramble. Ines and Bramble are made up for the show. Rob has built and broken these systems, and he explains them with a restaurant kitchen, a delivery truck and a conveyor belt. Over twelve episodes they ask which hurts more, late or wrong, with Uber and DoorDash, set Netflix beside a phone bill and a hospital beside a bank, name the stations behind a wall of logos, weigh the nightly truck against the live belt, and work through the lakehouse and data contracts, one written recipe for revenue, data that quietly stops arriving, checks from bronze to gold, a table nobody owns, and what AI that writes queries changes about the job. No technical background needed. More at cloudadorn.com.</description>
<itunes:summary>Somebody asks for revenue by region for the last three years, and it turns out the answer can&apos;t come straight from the tills. Modern Data Engineering is about the work behind answers like that: copying data from where it is born to where it is useful without breaking, losing or leaking it. Shelley has a boss who wants every dashboard live and a neighbour, Ines, who is a data engineer at a grocery chain called Bramble. Ines and Bramble are made up for the show. Rob has built and broken these systems, and he explains them with a restaurant kitchen, a delivery truck and a conveyor belt. Over twelve episodes they ask which hurts more, late or wrong, with Uber and DoorDash, set Netflix beside a phone bill and a hospital beside a bank, name the stations behind a wall of logos, weigh the nightly truck against the live belt, and work through the lakehouse and data contracts, one written recipe for revenue, data that quietly stops arriving, checks from bronze to gold, a table nobody owns, and what AI that writes queries changes about the job. No technical background needed. More at cloudadorn.com.</itunes:summary>
<itunes:subtitle>How real systems move data from where it is created to where it is useful.</itunes:subtitle>
<itunes:keywords>data engineering, data pipeline, data warehouse, data lake, streaming vs batch, dbt, Apache Airflow, Apache Kafka, lakehouse, data governance</itunes:keywords>
<link>https://www.cloudadorn.com/</link>
<language>en-us</language>
<copyright>Cloudadorn Academy</copyright>
<atom:link href="https://media.cloudadorn.org/modern-data-engineering/feed.xml" rel="self" type="application/rss+xml"/>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:category text="Technology"/>
<itunes:category text="Education">
<itunes:category text="Courses"/>
</itunes:category>
<itunes:explicit>false</itunes:explicit>
<itunes:author>Shelley and Rob</itunes:author>
<itunes:owner>
<itunes:name>Cloudadorn Academy</itunes:name>
<itunes:email>info@cloudadorn.com</itunes:email>
</itunes:owner>
<itunes:type>serial</itunes:type>
<itunes:complete>yes</itunes:complete>
<podcast:guid>5b3de2dc-73a4-51cb-8af2-4bdfbcdf5bbc</podcast:guid>
<podcast:locked owner="info@cloudadorn.com">no</podcast:locked>
<lastBuildDate>Sun, 16 Aug 2026 09:00:00 +0000</lastBuildDate>
<content:encoded><![CDATA[<p>Somebody asks for revenue by region for the last three years, and it turns out the answer can't come straight from the tills. Modern Data Engineering is about the work behind answers like that: copying data from where it is born to where it is useful without breaking, losing or leaking it. Shelley has a boss who wants every dashboard live and a neighbour, Ines, who is a data engineer at a grocery chain called Bramble. Ines and Bramble are made up for the show. Rob has built and broken these systems, and he explains them with a restaurant kitchen, a delivery truck and a conveyor belt. Over twelve episodes they ask which hurts more, late or wrong, with Uber and DoorDash, set Netflix beside a phone bill and a hospital beside a bank, name the stations behind a wall of logos, weigh the nightly truck against the live belt, and work through the lakehouse and data contracts, one written recipe for revenue, data that quietly stops arriving, checks from bronze to gold, a table nobody owns, and what AI that writes queries changes about the job. No technical background needed. More at cloudadorn.com.</p>]]></content:encoded>
<item>
<title>S1E01. What a Data Engineer Actually Does</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e01</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-01.mp3" length="15078654" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:00:00 +0000</pubDate>
<itunes:duration>583</itunes:duration>
<itunes:episode>1</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>data engineering, data lake, data warehouse, pipeline, checkout system, revenue by region, streaming</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-01.txt" type="text/plain"/>
<description>Shelley&apos;s boss wants revenue by region for the last three years, and the numbers are already in the checkout system, so why can&apos;t an analyst just look? Her neighbour Ines, a data engineer at a grocery chain called Bramble, says running that on Bramble&apos;s tills on a Saturday morning would push the queue into the car park. Ines and Bramble are both made up for the show. Rob swaps the grocery for a restaurant in the lunch rush and explains why the till system is built for recording sales, not reading them, and why the history gets copied, usually overnight, to a second store. You&apos;ll hear what a data lake and a data warehouse are, why a renamed column can make a report claim Bramble sold nothing on Tuesday, and what fills most of a data engineer&apos;s week. Next, a concert crowd and a fare that jumped within a minute.</description>
<itunes:summary>Shelley&apos;s boss wants revenue by region for the last three years, and the numbers are already in the checkout system, so why can&apos;t an analyst just look? Her neighbour Ines, a data engineer at a grocery chain called Bramble, says running that on Bramble&apos;s tills on a Saturday morning would push the queue into the car park. Ines and Bramble are both made up for the show. Rob swaps the grocery for a restaurant in the lunch rush and explains why the till system is built for recording sales, not reading them, and why the history gets copied, usually overnight, to a second store. You&apos;ll hear what a data lake and a data warehouse are, why a renamed column can make a report claim Bramble sold nothing on Tuesday, and what fills most of a data engineer&apos;s week. Next, a concert crowd and a fare that jumped within a minute.</itunes:summary>
<content:encoded><![CDATA[<p>Shelley's boss wants revenue by region for the last three years, and the numbers are already in the checkout system, so why can't an analyst just look? Her neighbour Ines, a data engineer at a grocery chain called Bramble, says running that on Bramble's tills on a Saturday morning would push the queue into the car park. Ines and Bramble are both made up for the show. Rob swaps the grocery for a restaurant in the lunch rush and explains why the till system is built for recording sales, not reading them, and why the history gets copied, usually overnight, to a second store. You'll hear what a data lake and a data warehouse are, why a renamed column can make a report claim Bramble sold nothing on Tuesday, and what fills most of a data engineer's week. Next, a concert crowd and a fare that jumped within a minute.</p>]]></content:encoded>
</item>
<item>
<title>S1E02. Late vs Wrong: When to Stream and When to Batch</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e02</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-02.mp3" length="13932756" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:02:00 +0000</pubDate>
<itunes:duration>535</itunes:duration>
<itunes:episode>2</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>Uber, DoorDash, Kafka, Flink, surge prices, streaming, batch, hexagons</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-02.txt" type="text/plain"/>
<description>Within a minute of a concert letting out, the price of a ride home had already jumped, and Shelley&apos;s friend has the Uber receipt to prove it. Ines, a neighbour the hosts invented along with Bramble, her grocery chain, says Bramble&apos;s supplier payments never move that fast. They run overnight and get checked before a penny moves. Rob says both clocks are right, and the question that picks one is which hurts more, late or a little wrong. He walks through what Uber&apos;s engineers wrote in 2018 about laying hexagons over a city to set surge prices, and in 2020 about Kafka, a conveyor belt for messages, read by programs built on Flink. DoorDash&apos;s 2022 conference talk is the second witness. You&apos;ll understand why the driver&apos;s pay waits for a checked batch, and how a belt keeps one trip&apos;s messages in order. Next, a paused show and a phone bill.</description>
<itunes:summary>Within a minute of a concert letting out, the price of a ride home had already jumped, and Shelley&apos;s friend has the Uber receipt to prove it. Ines, a neighbour the hosts invented along with Bramble, her grocery chain, says Bramble&apos;s supplier payments never move that fast. They run overnight and get checked before a penny moves. Rob says both clocks are right, and the question that picks one is which hurts more, late or a little wrong. He walks through what Uber&apos;s engineers wrote in 2018 about laying hexagons over a city to set surge prices, and in 2020 about Kafka, a conveyor belt for messages, read by programs built on Flink. DoorDash&apos;s 2022 conference talk is the second witness. You&apos;ll understand why the driver&apos;s pay waits for a checked batch, and how a belt keeps one trip&apos;s messages in order. Next, a paused show and a phone bill.</itunes:summary>
<content:encoded><![CDATA[<p>Within a minute of a concert letting out, the price of a ride home had already jumped, and Shelley's friend has the Uber receipt to prove it. Ines, a neighbour the hosts invented along with Bramble, her grocery chain, says Bramble's supplier payments never move that fast. They run overnight and get checked before a penny moves. Rob says both clocks are right, and the question that picks one is which hurts more, late or a little wrong. He walks through what Uber's engineers wrote in 2018 about laying hexagons over a city to set surge prices, and in 2020 about Kafka, a conveyor belt for messages, read by programs built on Flink. DoorDash's 2022 conference talk is the second witness. You'll understand why the driver's pay waits for a checked batch, and how a belt keeps one trip's messages in order. Next, a paused show and a phone bill.</p>]]></content:encoded>
</item>
<item>
<title>S1E03. Netflix and Telecom: Same Data, Different Fear</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e03</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-03.mp3" length="13235506" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:04:00 +0000</pubDate>
<itunes:duration>506</itunes:duration>
<itunes:episode>3</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>Netflix, Kafka, call record, phone bill, duplicates, three-way check, home screen</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-03.txt" type="text/plain"/>
<description>Shelley paused a show on the television on Sunday, and on Monday her phone picked it up at the exact minute. On Wednesday her phone bill listed one call to her sister twice. Rob says each company is frightened of a different mistake. Netflix appears on Kafka&apos;s own list of the companies that use it, for watching things live, and Netflix&apos;s help page explains how a home screen gets worked out. Then Rob follows the call record, through American rules that make long distance billers keep those records for eighteen months and protect your location, to how duplicates get thrown out before anything is priced. Shelley&apos;s neighbour Ines spends her week on duplicate till receipts at Bramble, a grocer as made up as she is. A monthly bill may take all night, but it must not be wrong. Coming up, two copies of Shelley&apos;s dad.</description>
<itunes:summary>Shelley paused a show on the television on Sunday, and on Monday her phone picked it up at the exact minute. On Wednesday her phone bill listed one call to her sister twice. Rob says each company is frightened of a different mistake. Netflix appears on Kafka&apos;s own list of the companies that use it, for watching things live, and Netflix&apos;s help page explains how a home screen gets worked out. Then Rob follows the call record, through American rules that make long distance billers keep those records for eighteen months and protect your location, to how duplicates get thrown out before anything is priced. Shelley&apos;s neighbour Ines spends her week on duplicate till receipts at Bramble, a grocer as made up as she is. A monthly bill may take all night, but it must not be wrong. Coming up, two copies of Shelley&apos;s dad.</itunes:summary>
<content:encoded><![CDATA[<p>Shelley paused a show on the television on Sunday, and on Monday her phone picked it up at the exact minute. On Wednesday her phone bill listed one call to her sister twice. Rob says each company is frightened of a different mistake. Netflix appears on Kafka's own list of the companies that use it, for watching things live, and Netflix's help page explains how a home screen gets worked out. Then Rob follows the call record, through American rules that make long distance billers keep those records for eighteen months and protect your location, to how duplicates get thrown out before anything is priced. Shelley's neighbour Ines spends her week on duplicate till receipts at Bramble, a grocer as made up as she is. A monthly bill may take all night, but it must not be wrong. Coming up, two copies of Shelley's dad.</p>]]></content:encoded>
</item>
<item>
<title>S1E04. Hospital and Bank Data: Privacy, Fraud, Regulators</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e04</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-04.mp3" length="13035345" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:06:00 +0000</pubDate>
<itunes:duration>498</itunes:duration>
<itunes:episode>4</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>HIPAA, master patient index, know your customer, money laundering, patient portal, token, Jane Smith</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-04.txt" type="text/plain"/>
<description>After Shelley&apos;s dad spent Tuesday night in the emergency room, the hospital&apos;s patient portal showed two of him, one with a letter missing from his surname and the X-ray under the wrong one. On Friday a bank wanted passports and the real owners of her pottery stall, which is two women and a kiln. Rob explains why the hospital fears joining two different people while the bank fears missing one person hiding behind six names. You&apos;ll hear how a master patient index decides when three rows are one human, what HIPAA, the American health privacy law, says about eighteen kinds of identifying detail and breaches of more than five hundred people, what know your customer means, and why a tapped card becomes a token. Ines, the neighbour, and Bramble, the grocery chain she works for, are both invented. The next one opens on a blank commission report.</description>
<itunes:summary>After Shelley&apos;s dad spent Tuesday night in the emergency room, the hospital&apos;s patient portal showed two of him, one with a letter missing from his surname and the X-ray under the wrong one. On Friday a bank wanted passports and the real owners of her pottery stall, which is two women and a kiln. Rob explains why the hospital fears joining two different people while the bank fears missing one person hiding behind six names. You&apos;ll hear how a master patient index decides when three rows are one human, what HIPAA, the American health privacy law, says about eighteen kinds of identifying detail and breaches of more than five hundred people, what know your customer means, and why a tapped card becomes a token. Ines, the neighbour, and Bramble, the grocery chain she works for, are both invented. The next one opens on a blank commission report.</itunes:summary>
<content:encoded><![CDATA[<p>After Shelley's dad spent Tuesday night in the emergency room, the hospital's patient portal showed two of him, one with a letter missing from his surname and the X-ray under the wrong one. On Friday a bank wanted passports and the real owners of her pottery stall, which is two women and a kiln. Rob explains why the hospital fears joining two different people while the bank fears missing one person hiding behind six names. You'll hear how a master patient index decides when three rows are one human, what HIPAA, the American health privacy law, says about eighteen kinds of identifying detail and breaches of more than five hundred people, what know your customer means, and why a tapped card becomes a token. Ines, the neighbour, and Bramble, the grocery chain she works for, are both invented. The next one opens on a blank commission report.</p>]]></content:encoded>
</item>
<item>
<title>S1E05. The Data Stack Explained as a Kitchen Line</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e05</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-05.mp3" length="13738149" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:08:00 +0000</pubDate>
<itunes:duration>527</itunes:duration>
<itunes:episode>5</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>Salesforce, change data capture, data lake, data warehouse, Snowflake, Spark, dbt, Airflow</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-05.txt" type="text/plain"/>
<description>On Monday the report telling Shelley&apos;s sales reps what they&apos;re owed came up blank, the data team said the platform was down, and the only map was a slide full of logos. Her neighbour Ines, a data engineer at Bramble, a grocery chain made up along with Ines herself, would ask which station broke. So Rob draws the kitchen behind the logos: a farm where the stock comes from, a truck that brings it, and a schedule board on the wall. You&apos;ll meet Salesforce&apos;s limit on requests per day, change data capture and why Amazon&apos;s own instructions say it isn&apos;t real time, the data lake and the data warehouse, why Snowflake uses the word warehouse for the cook, and what Spark, dbt and Airflow each do. Then Shelley learns what happened to Sunday&apos;s copy. Next, a boss who wants it live by Monday.</description>
<itunes:summary>On Monday the report telling Shelley&apos;s sales reps what they&apos;re owed came up blank, the data team said the platform was down, and the only map was a slide full of logos. Her neighbour Ines, a data engineer at Bramble, a grocery chain made up along with Ines herself, would ask which station broke. So Rob draws the kitchen behind the logos: a farm where the stock comes from, a truck that brings it, and a schedule board on the wall. You&apos;ll meet Salesforce&apos;s limit on requests per day, change data capture and why Amazon&apos;s own instructions say it isn&apos;t real time, the data lake and the data warehouse, why Snowflake uses the word warehouse for the cook, and what Spark, dbt and Airflow each do. Then Shelley learns what happened to Sunday&apos;s copy. Next, a boss who wants it live by Monday.</itunes:summary>
<content:encoded><![CDATA[<p>On Monday the report telling Shelley's sales reps what they're owed came up blank, the data team said the platform was down, and the only map was a slide full of logos. Her neighbour Ines, a data engineer at Bramble, a grocery chain made up along with Ines herself, would ask which station broke. So Rob draws the kitchen behind the logos: a farm where the stock comes from, a truck that brings it, and a schedule board on the wall. You'll meet Salesforce's limit on requests per day, change data capture and why Amazon's own instructions say it isn't real time, the data lake and the data warehouse, why Snowflake uses the word warehouse for the cook, and what Spark, dbt and Airflow each do. Then Shelley learns what happened to Sunday's copy. Next, a boss who wants it live by Monday.</p>]]></content:encoded>
</item>
<item>
<title>S1E06. Batch vs Streaming: Why Nightly Is Usually Right</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e06</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-06.mp3" length="15371115" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:10:00 +0000</pubDate>
<itunes:duration>595</itunes:duration>
<itunes:episode>6</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>batch, streaming, change data capture, watermark, Kafka, at least once, exactly once, live dashboard</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-06.txt" type="text/plain"/>
<description>By Monday, Shelley&apos;s boss wants the sales dashboard live, and she has a weekend to judge whether that&apos;s reasonable. Ines next door, who runs the data side at Bramble, has two stories from one Tuesday: finance reading yesterday&apos;s sales at eight in the morning, and a stolen card tried over and over on Bramble&apos;s website at a quarter past two. Ines and her grocer are made up, but Rob says those stories turn up in every real business. With a delivery truck, a shuttle van and a conveyor belt, he shows how batch, little batches on a fast schedule, and streaming differ. You&apos;ll understand why grabbing rows by date misses deletions, what a watermark does for a phone tap stuck in a tunnel, why Kafka promises at least once, and what to ask when Google promises exactly once. Up next, a conference slide with two words on it.</description>
<itunes:summary>By Monday, Shelley&apos;s boss wants the sales dashboard live, and she has a weekend to judge whether that&apos;s reasonable. Ines next door, who runs the data side at Bramble, has two stories from one Tuesday: finance reading yesterday&apos;s sales at eight in the morning, and a stolen card tried over and over on Bramble&apos;s website at a quarter past two. Ines and her grocer are made up, but Rob says those stories turn up in every real business. With a delivery truck, a shuttle van and a conveyor belt, he shows how batch, little batches on a fast schedule, and streaming differ. You&apos;ll understand why grabbing rows by date misses deletions, what a watermark does for a phone tap stuck in a tunnel, why Kafka promises at least once, and what to ask when Google promises exactly once. Up next, a conference slide with two words on it.</itunes:summary>
<content:encoded><![CDATA[<p>By Monday, Shelley's boss wants the sales dashboard live, and she has a weekend to judge whether that's reasonable. Ines next door, who runs the data side at Bramble, has two stories from one Tuesday: finance reading yesterday's sales at eight in the morning, and a stolen card tried over and over on Bramble's website at a quarter past two. Ines and her grocer are made up, but Rob says those stories turn up in every real business. With a delivery truck, a shuttle van and a conveyor belt, he shows how batch, little batches on a fast schedule, and streaming differ. You'll understand why grabbing rows by date misses deletions, what a watermark does for a phone tap stuck in a tunnel, why Kafka promises at least once, and what to ask when Google promises exactly once. Up next, a conference slide with two words on it.</p>]]></content:encoded>
</item>
<item>
<title>S1E07. Lakehouse Explained: Iceberg, Contracts, and Mesh</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e07</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-07.mp3" length="14522609" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:12:00 +0000</pubDate>
<itunes:duration>560</itunes:duration>
<itunes:episode>7</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>lakehouse, data mesh, Iceberg, Netflix, data contract, walk-in fridge, warehouse</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-07.txt" type="text/plain"/>
<description>Lakehouse across the top, data mesh underneath: that&apos;s the slide Shelley&apos;s boss brought back from a conference, and he wants a plan for both by Friday. Ines, who looks after the data at Bramble, a grocery chain the hosts made up along with her, spent last winter with two copies of the sales data that kept disagreeing. Rob explains the lakehouse as a cheap walk-in fridge of files with a clipboard on the door listing exactly which files make up a table, so a bad delivery can be undone. You&apos;ll hear why Iceberg, first built inside Netflix, is an open project no single company owns, how a data contract stops a renamed column before it goes live, and why data mesh was written for very large companies rather than a team of two analysts. Next, three revenues in one Monday meeting.</description>
<itunes:summary>Lakehouse across the top, data mesh underneath: that&apos;s the slide Shelley&apos;s boss brought back from a conference, and he wants a plan for both by Friday. Ines, who looks after the data at Bramble, a grocery chain the hosts made up along with her, spent last winter with two copies of the sales data that kept disagreeing. Rob explains the lakehouse as a cheap walk-in fridge of files with a clipboard on the door listing exactly which files make up a table, so a bad delivery can be undone. You&apos;ll hear why Iceberg, first built inside Netflix, is an open project no single company owns, how a data contract stops a renamed column before it goes live, and why data mesh was written for very large companies rather than a team of two analysts. Next, three revenues in one Monday meeting.</itunes:summary>
<content:encoded><![CDATA[<p>Lakehouse across the top, data mesh underneath: that's the slide Shelley's boss brought back from a conference, and he wants a plan for both by Friday. Ines, who looks after the data at Bramble, a grocery chain the hosts made up along with her, spent last winter with two copies of the sales data that kept disagreeing. Rob explains the lakehouse as a cheap walk-in fridge of files with a clipboard on the door listing exactly which files make up a table, so a bad delivery can be undone. You'll hear why Iceberg, first built inside Netflix, is an open project no single company owns, how a data contract stops a renamed column before it goes live, and why data mesh was written for very large companies rather than a team of two analysts. Next, three revenues in one Monday meeting.</p>]]></content:encoded>
</item>
<item>
<title>S1E08. Metrics, Feature Stores, and Sending Data Back</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e08</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-08.mp3" length="14408432" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:14:00 +0000</pubDate>
<itunes:duration>555</itunes:duration>
<itunes:episode>8</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>star schema, dimensions, slowly changing dimension, metrics layer, Power BI, semantic model, reverse ETL, Salesforce</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-08.txt" type="text/plain"/>
<description>Finance, marketing and Shelley&apos;s boss each showed a different revenue for the same quarter on Monday, and he wants the real one by Wednesday. Ines says Bramble, the grocery chain where she&apos;s a data engineer, had the same fight for a year, and both were invented for the show. Rob starts with the layout, a long pile of restaurant ticket lines with small books beside it, called a star schema. With the guides from Ralph Kimball&apos;s group, he explains why deciding what one row means is the pivotal step, how a shopper who moved from Boston to Denver took her old sales with her, and the fix called a slowly changing dimension. Then come one recipe per number, the metrics layer, Power BI&apos;s semantic model, and reverse ETL, which puts a warning score in front of Lena in Salesforce. Next, a weekend sale that seemed to do nothing.</description>
<itunes:summary>Finance, marketing and Shelley&apos;s boss each showed a different revenue for the same quarter on Monday, and he wants the real one by Wednesday. Ines says Bramble, the grocery chain where she&apos;s a data engineer, had the same fight for a year, and both were invented for the show. Rob starts with the layout, a long pile of restaurant ticket lines with small books beside it, called a star schema. With the guides from Ralph Kimball&apos;s group, he explains why deciding what one row means is the pivotal step, how a shopper who moved from Boston to Denver took her old sales with her, and the fix called a slowly changing dimension. Then come one recipe per number, the metrics layer, Power BI&apos;s semantic model, and reverse ETL, which puts a warning score in front of Lena in Salesforce. Next, a weekend sale that seemed to do nothing.</itunes:summary>
<content:encoded><![CDATA[<p>Finance, marketing and Shelley's boss each showed a different revenue for the same quarter on Monday, and he wants the real one by Wednesday. Ines says Bramble, the grocery chain where she's a data engineer, had the same fight for a year, and both were invented for the show. Rob starts with the layout, a long pile of restaurant ticket lines with small books beside it, called a star schema. With the guides from Ralph Kimball's group, he explains why deciding what one row means is the pivotal step, how a shopper who moved from Boston to Denver took her old sales with her, and the fix called a slowly changing dimension. Then come one recipe per number, the metrics layer, Power BI's semantic model, and reverse ETL, which puts a warning score in front of Lena in Salesforce. Next, a weekend sale that seemed to do nothing.</p>]]></content:encoded>
</item>
<item>
<title>S1E09. DataOps and FinOps: Tests, Alerts, and the Bill</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e09</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-09.mp3" length="12800676" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:16:00 +0000</pubDate>
<itunes:duration>488</itunes:duration>
<itunes:episode>9</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>data observability, freshness check, lineage, Monte Carlo, Airflow, Snowflake, DataOps, monitoring</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-09.txt" type="text/plain"/>
<description>Shelley&apos;s free delivery weekend looked like a flop on Monday, the dashboard all green and her boss telling the team it hadn&apos;t done much. Then she noticed the newest order on it was from Thursday afternoon. Nothing went red, Rob explains, because the nightly rebuild ran fine and made a perfect copy of Thursday. You&apos;ll learn the difference between checking that a job ran and watching the data itself, starting with a freshness check, the date on the milk, and what vendors such as Monte Carlo mean by data observability. Lineage, walked backwards, finds the stuck copy and, walked forwards, finds finance&apos;s weekly report. Airflow assets start work when data arrives, Snowflake may explain some of the bill, and DataOps gets a plain meaning. Ines, the neighbour who talked Shelley&apos;s team into tests, is made up, and so is Bramble. After that comes a correction email about the northern stores.</description>
<itunes:summary>Shelley&apos;s free delivery weekend looked like a flop on Monday, the dashboard all green and her boss telling the team it hadn&apos;t done much. Then she noticed the newest order on it was from Thursday afternoon. Nothing went red, Rob explains, because the nightly rebuild ran fine and made a perfect copy of Thursday. You&apos;ll learn the difference between checking that a job ran and watching the data itself, starting with a freshness check, the date on the milk, and what vendors such as Monte Carlo mean by data observability. Lineage, walked backwards, finds the stuck copy and, walked forwards, finds finance&apos;s weekly report. Airflow assets start work when data arrives, Snowflake may explain some of the bill, and DataOps gets a plain meaning. Ines, the neighbour who talked Shelley&apos;s team into tests, is made up, and so is Bramble. After that comes a correction email about the northern stores.</itunes:summary>
<content:encoded><![CDATA[<p>Shelley's free delivery weekend looked like a flop on Monday, the dashboard all green and her boss telling the team it hadn't done much. Then she noticed the newest order on it was from Thursday afternoon. Nothing went red, Rob explains, because the nightly rebuild ran fine and made a perfect copy of Thursday. You'll learn the difference between checking that a job ran and watching the data itself, starting with a freshness check, the date on the milk, and what vendors such as Monte Carlo mean by data observability. Lineage, walked backwards, finds the stuck copy and, walked forwards, finds finance's weekly report. Airflow assets start work when data arrives, Snowflake may explain some of the bill, and DataOps gets a plain meaning. Ines, the neighbour who talked Shelley's team into tests, is made up, and so is Bramble. After that comes a correction email about the northern stores.</p>]]></content:encoded>
</item>
<item>
<title>S1E10. Data Quality Gates: Bronze, Silver, and Gold</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e10</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-10.mp3" length="15017806" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:18:00 +0000</pubDate>
<itunes:duration>580</itunes:duration>
<itunes:episode>10</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>data quality, bronze, silver, gold, Databricks, dbt, dead-letter queue, conveyor belt</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-10.txt" type="text/plain"/>
<description>Correction to last week&apos;s northern stores figures. That was the subject line of the email Shelley&apos;s boss sent to twenty people after a believable number on his slide turned out to be missing a chunk of orders. Now he&apos;s read that data quality is one line of configuration and wants it by Friday. Ines, every bit as imaginary as Bramble, the grocery chain she works for, says it depends which shelf. Rob lays out bronze, silver and gold, a layout Databricks describes as improving quality at each step, with a checked door between them. You&apos;ll hear what the one line in dbt does, why someone who knows the business must choose the columns, how comparing with the same weekday catches a light delivery, why gold gets checked against the till to the penny, and what a dead-letter queue is for. Next, a blank column and a table nobody owns.</description>
<itunes:summary>Correction to last week&apos;s northern stores figures. That was the subject line of the email Shelley&apos;s boss sent to twenty people after a believable number on his slide turned out to be missing a chunk of orders. Now he&apos;s read that data quality is one line of configuration and wants it by Friday. Ines, every bit as imaginary as Bramble, the grocery chain she works for, says it depends which shelf. Rob lays out bronze, silver and gold, a layout Databricks describes as improving quality at each step, with a checked door between them. You&apos;ll hear what the one line in dbt does, why someone who knows the business must choose the columns, how comparing with the same weekday catches a light delivery, why gold gets checked against the till to the penny, and what a dead-letter queue is for. Next, a blank column and a table nobody owns.</itunes:summary>
<content:encoded><![CDATA[<p>Correction to last week's northern stores figures. That was the subject line of the email Shelley's boss sent to twenty people after a believable number on his slide turned out to be missing a chunk of orders. Now he's read that data quality is one line of configuration and wants it by Friday. Ines, every bit as imaginary as Bramble, the grocery chain she works for, says it depends which shelf. Rob lays out bronze, silver and gold, a layout Databricks describes as improving quality at each step, with a checked door between them. You'll hear what the one line in dbt does, why someone who knows the business must choose the columns, how comparing with the same weekday catches a light delivery, why gold gets checked against the till to the penny, and what a dead-letter queue is for. Next, a blank column and a table nobody owns.</p>]]></content:encoded>
</item>
<item>
<title>S1E11. Data Governance: Owners, Catalogs, and Lineage</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e11</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-11.mp3" length="13647174" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:20:00 +0000</pubDate>
<itunes:duration>523</itunes:duration>
<itunes:episode>11</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>governance, catalog, lineage, masking, GDPR, time travel, Snowflake, roles</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-11.txt" type="text/plain"/>
<description>At Tuesday&apos;s quarterly review the customer segment column on the revenue dashboard is blank, and when the chief financial officer asks who owns the table, someone says Maya built it. Maya left in March. Now it&apos;s Shelley&apos;s problem, by Friday. Rob says governance mostly means knowing what data you have, who is responsible for each piece and who is allowed to see it. Ines, Bramble&apos;s smug data engineer, is imagined like the grocery chain, but her rule about owners is worth stealing. You&apos;ll hear what a catalog can and can&apos;t fill in, how lineage finds where the segment stopped, and why access goes to roles, not to Owen, a keen new hire. Then masking, with Microsoft&apos;s own warning, and a GDPR deletion request that Snowflake&apos;s time travel complicates. Next, a chart drawn by a chat box at a vendor lunch.</description>
<itunes:summary>At Tuesday&apos;s quarterly review the customer segment column on the revenue dashboard is blank, and when the chief financial officer asks who owns the table, someone says Maya built it. Maya left in March. Now it&apos;s Shelley&apos;s problem, by Friday. Rob says governance mostly means knowing what data you have, who is responsible for each piece and who is allowed to see it. Ines, Bramble&apos;s smug data engineer, is imagined like the grocery chain, but her rule about owners is worth stealing. You&apos;ll hear what a catalog can and can&apos;t fill in, how lineage finds where the segment stopped, and why access goes to roles, not to Owen, a keen new hire. Then masking, with Microsoft&apos;s own warning, and a GDPR deletion request that Snowflake&apos;s time travel complicates. Next, a chart drawn by a chat box at a vendor lunch.</itunes:summary>
<content:encoded><![CDATA[<p>At Tuesday's quarterly review the customer segment column on the revenue dashboard is blank, and when the chief financial officer asks who owns the table, someone says Maya built it. Maya left in March. Now it's Shelley's problem, by Friday. Rob says governance mostly means knowing what data you have, who is responsible for each piece and who is allowed to see it. Ines, Bramble's smug data engineer, is imagined like the grocery chain, but her rule about owners is worth stealing. You'll hear what a catalog can and can't fill in, how lineage finds where the segment stopped, and why access goes to roles, not to Owen, a keen new hire. Then masking, with Microsoft's own warning, and a GDPR deletion request that Snowflake's time travel complicates. Next, a chart drawn by a chat box at a vendor lunch.</p>]]></content:encoded>
</item>
<item>
<title>S1E12. How Generative AI Changes Data Engineering</title>
<guid isPermaLink="false">cloudadorn.academy.mde.s01e12</guid>
<enclosure url="https://media.cloudadorn.org/modern-data-engineering/audio/episode-12.mp3" length="12107132" type="audio/mpeg"/>
<pubDate>Sun, 16 Aug 2026 09:22:00 +0000</pubDate>
<itunes:duration>459</itunes:duration>
<itunes:episode>12</itunes:episode>
<itunes:season>1</itunes:season>
<itunes:episodeType>full</itunes:episodeType>
<itunes:explicit>false</itunes:explicit>
<itunes:image href="https://media.cloudadorn.org/modern-data-engineering/art/show-cover-3000.jpg"/>
<itunes:keywords>chat box, Redshift, Gemini, BigQuery, hallucinations, prompt injection, agent, table descriptions</itunes:keywords>
<podcast:transcript url="https://media.cloudadorn.org/modern-data-engineering/transcripts/episode-12.txt" type="text/plain"/>
<description>Show me revenue by region for last quarter. Someone typed that into a chat box at a vendor lunch, it drew Shelley&apos;s boss a chart before he&apos;d finished his coffee, and he asked why the company had just advertised for a data engineer. Rob pictures the model as a new cook, fast, never tired and happy to serve whatever was left on the counter. Amazon&apos;s page for the query assistant in its Redshift warehouse and Google&apos;s page for Gemini in BigQuery both tell you to check what it writes. You&apos;ll understand why table descriptions suddenly matter, how Ines, a data engineer at Bramble, both of them pretend, guards against a seven on a scanned delivery note being read as a one, what prompt injection is, and why an agent needs its own login. Shelley answers her boss by text from the car park. More at cloudadorn.com.</description>
<itunes:summary>Show me revenue by region for last quarter. Someone typed that into a chat box at a vendor lunch, it drew Shelley&apos;s boss a chart before he&apos;d finished his coffee, and he asked why the company had just advertised for a data engineer. Rob pictures the model as a new cook, fast, never tired and happy to serve whatever was left on the counter. Amazon&apos;s page for the query assistant in its Redshift warehouse and Google&apos;s page for Gemini in BigQuery both tell you to check what it writes. You&apos;ll understand why table descriptions suddenly matter, how Ines, a data engineer at Bramble, both of them pretend, guards against a seven on a scanned delivery note being read as a one, what prompt injection is, and why an agent needs its own login. Shelley answers her boss by text from the car park. More at cloudadorn.com.</itunes:summary>
<content:encoded><![CDATA[<p>Show me revenue by region for last quarter. Someone typed that into a chat box at a vendor lunch, it drew Shelley's boss a chart before he'd finished his coffee, and he asked why the company had just advertised for a data engineer. Rob pictures the model as a new cook, fast, never tired and happy to serve whatever was left on the counter. Amazon's page for the query assistant in its Redshift warehouse and Google's page for Gemini in BigQuery both tell you to check what it writes. You'll understand why table descriptions suddenly matter, how Ines, a data engineer at Bramble, both of them pretend, guards against a seven on a scanned delivery note being read as a one, what prompt injection is, and why an agent needs its own login. Shelley answers her boss by text from the car park. More at cloudadorn.com.</p>]]></content:encoded>
</item>
</channel></rss>
