<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Paul Rossman on Medium]]></title>
        <description><![CDATA[Stories by Paul Rossman on Medium]]></description>
        <link>https://medium.com/@psrossman?source=rss-6903991ebf07------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*Papi5gltNOun9rwk.jpg</url>
            <title>Stories by Paul Rossman on Medium</title>
            <link>https://medium.com/@psrossman?source=rss-6903991ebf07------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 18:35:01 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@psrossman/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Converting a Dashboard from Tableau to Databricks AI/BI using Claude]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@psrossman/converting-a-dashboard-from-tableau-to-databricks-ai-bi-using-claude-16778e4bded6?source=rss-6903991ebf07------2"><img src="https://cdn-images-1.medium.com/max/1715/1*Oemo3no2kPFMmlSH73eX_Q.png" width="1715"></a></p><p class="medium-feed-snippet">Recently, I have been &#x2018;vibeboarding&#x2019;&#x200A;&#x2014;&#x200A;that is vibecoding a dashboard. In this article, I&#x2019;m going to give a walkthrough of how I converted&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@psrossman/converting-a-dashboard-from-tableau-to-databricks-ai-bi-using-claude-16778e4bded6?source=rss-6903991ebf07------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@psrossman/converting-a-dashboard-from-tableau-to-databricks-ai-bi-using-claude-16778e4bded6?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/16778e4bded6</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[business-intelligence]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Mon, 29 Dec 2025 13:16:50 GMT</pubDate>
            <atom:updated>2025-12-29T13:16:50.653Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Building a Custom Export for AI/BI Dashboards]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@psrossman/building-a-custom-export-for-ai-bi-dashboards-4e5be02b0178?source=rss-6903991ebf07------2"><img src="https://cdn-images-1.medium.com/max/752/1*FqgOumEwSB4DS5I_VfgZAg.png" width="752"></a></p><p class="medium-feed-snippet">Hello friends &#x2014;</p><p class="medium-feed-link"><a href="https://medium.com/@psrossman/building-a-custom-export-for-ai-bi-dashboards-4e5be02b0178?source=rss-6903991ebf07------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@psrossman/building-a-custom-export-for-ai-bi-dashboards-4e5be02b0178?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/4e5be02b0178</guid>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sat, 13 Dec 2025 17:29:47 GMT</pubDate>
            <atom:updated>2025-12-13T17:29:47.921Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Creating a Bullet Chart in AI/BI Dashboards]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@psrossman/creating-a-bullet-chart-in-ai-bi-dashboards-9be0eadffa29?source=rss-6903991ebf07------2"><img src="https://cdn-images-1.medium.com/max/1148/1*qJrnq4ukdPd-jOoX4ZLmSg.png" width="1148"></a></p><p class="medium-feed-snippet">Hello friends, welcome to another post around AI/BI Dashboards. In this one we will create a bullet chart, since this is not one of the&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@psrossman/creating-a-bullet-chart-in-ai-bi-dashboards-9be0eadffa29?source=rss-6903991ebf07------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@psrossman/creating-a-bullet-chart-in-ai-bi-dashboards-9be0eadffa29?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/9be0eadffa29</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[dashboard]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sun, 12 Oct 2025 14:38:20 GMT</pubDate>
            <atom:updated>2025-10-12T14:38:20.494Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Using the new spatial SQL functions in Databricks (with ArcGIS data)]]></title>
            <link>https://medium.com/@psrossman/using-the-new-spatial-sql-functions-in-databricks-with-arcgis-data-563f4dd9f09f?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/563f4dd9f09f</guid>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[sql]]></category>
            <category><![CDATA[spatial]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sat, 06 Sep 2025 14:51:13 GMT</pubDate>
            <atom:updated>2025-09-06T14:51:13.372Z</atom:updated>
            <content:encoded><![CDATA[<p>This is a quick tutorial on how to pull down data from ArcGIS, store it in Unity catalog on Databricks, and then leverage the new spatial SQL functions to perform some analysis.</p><p><strong>Getting started</strong></p><p>First, let’s pull some data down from a feature layer from ArcGIS.</p><ol><li>If you don’t have an account, you can sign up <a href="https://www.esri.com/en-us/arcgis/products/create-account">here</a>.</li><li>Let’s login to the free edition of Databricks.</li><li>Install the python package for ArcGIS:</li></ol><pre>%pip install arcgis</pre><p>4. Let’s login to ArcGIS programmically:</p><pre>from arcgis import GIS<br><br>print(&quot;Portal for ArcGIS as a built in user&quot;)<br>gis = GIS(<br>    &quot;https://arcgis.com&quot;,<br>    username=&quot;&quot;,<br>    password=&quot;&quot;<br>)<br>print(&quot;Logged in as: &quot; + gis.properties.user.username)</pre><p>5. Let’s pull down a feature layer from ArcGIS Online.</p><p>I like to do a <a href="https://www.arcgis.com/home/search.html?restrict=false&amp;sortField=relevance&amp;sortOrder=desc&amp;focus=layers-weblayers-features#content">search</a> to find what I need. Let’s keep this simple and pull down the boundaries for US Counties. By clicking on view details, we find the item id, which we need for the API call:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BwaC62VlrSQBnQdByRRxvw.png" /></figure><p>Now, let’s write some code to pull it programmically:</p><pre>item = gis.content.get(&quot;3c164274a80748dda926a046525da610&quot;)<br>display(item)</pre><p>Next, let’s extract the feature layer and pull it into pandas:</p><pre>import pandas as pd<br>if hasattr(item, &quot;layers&quot;):<br>    flayer = item.layers[0]<br>else:<br>    print(&quot;This item does not have layers. Check the item type and use the appropriate property.&quot;)<br><br>sdf = pd.DataFrame.spatial.from_layer(flayer)<br>sdf</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nsLnC-O4nQkbAIcLudMYpA.png" /></figure><p><strong>Storing it in a table in Unity Catalog</strong></p><p>Before we write directly to unity, let’s find the coordinate reference system (CRS). For that, let’s install shapely and pyproj:</p><pre>!pip install shapely pyproj</pre><pre>crs = flayer.properties.spatialReference<br><br>display(crs)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/353/1*Zz3mxGCoSknZD83S6hQhIQ.png" /></figure><p>I always like to check this, that way if it’s in a CRS we don’t like we can transform it into one we do. <a href="https://epsg.io/4326">4326</a> is perfect for our needs.</p><p>Next, let’s convert the SHAPE field into a WKT (Well-Known Text). Otherwise, it will write it as a binary to the table.</p><pre>from shapely.geometry import shape<br><br>sdf[&#39;geometry_wkt&#39;] = sdf[&#39;SHAPE&#39;].apply(lambda x: shape(x.__geo_interface__).wkt)</pre><p>Finally, let’s write to Unity:</p><pre>sdf_spark = spark.createDataFrame(sdf)<br>sdf_spark.write.format(&quot;delta&quot;).mode(&quot;overwrite&quot;).saveAsTable(&quot;workspace.default.us_counties&quot;)</pre><p><strong>Using the new spatial SQL functions</strong></p><p>Now we can use this new table in spatial queries. Here are a couple examples:</p><p>Let’s find which county the bakehouse franchises are located in:</p><pre>%sql<br>select <br>    franchiseID, FIPS, usc.NAME, usc.STATE_NAME<br>from <br>    workspace.default.us_counties usc<br>inner join <br>    samples.bakehouse.sales_franchises store<br>on <br>    st_contains(st_geomfromtext(usc.geometry_wkt), st_point(store.Longitude, store.Latitude))<br>order by FIPS</pre><p>Let’s find the centroid of each county:</p><pre>%sql<br>SELECT fips, st_asewkt(st_centroid(st_geomfromtext(usc.geometry_wkt))) FROM workspace.default.us_counties usc;</pre><p>Find the closest bakeshop in miles to the center of San Fransisco, County:</p><pre>%sql<br>select franchiseID, name, city, st_distancespheroid(st_geomfromtext(&#39;POINT(-122.43010255822645 37.746700255675606)&#39;), st_point(Longitude, Latitude)) / 1609.34 as distance_miles<br>from samples.bakehouse.sales_franchises store<br>order by distance_miles asc</pre><p>These are only a few examples. Here is the <a href="https://docs.databricks.com/aws/en/sql/language-manual/sql-ref-st-geospatial-functions">full list</a> of built in spatial sql functions.</p><p>Thanks for reading!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=563f4dd9f09f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building Your First Metric View in Databricks (and Using it in a Dashboard)]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@psrossman/building-your-first-metric-view-in-databricks-and-using-it-in-a-dashboard-40d3f0dd50a5?source=rss-6903991ebf07------2"><img src="https://cdn-images-1.medium.com/max/600/1*j8nRc3GsrtCESZgVUbzfrA.png" width="600"></a></p><p class="medium-feed-snippet">Over the summer, metric views in Databricks became available, after being announced on stage at the 2024 Data and AI Summit. This post&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@psrossman/building-your-first-metric-view-in-databricks-and-using-it-in-a-dashboard-40d3f0dd50a5?source=rss-6903991ebf07------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@psrossman/building-your-first-metric-view-in-databricks-and-using-it-in-a-dashboard-40d3f0dd50a5?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/40d3f0dd50a5</guid>
            <category><![CDATA[semantic-layer]]></category>
            <category><![CDATA[databricks]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sun, 17 Aug 2025 12:29:49 GMT</pubDate>
            <atom:updated>2025-08-17T12:29:49.286Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Databricks Dashboards — Building KPI Indicators with HTML Styling]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@psrossman/databricks-dashboards-building-kpi-indicators-with-html-styling-78a9275d7f5d?source=rss-6903991ebf07------2"><img src="https://cdn-images-1.medium.com/max/1262/1*rIJzjr5x3dB_c3gNoZw4_w.png" width="1262"></a></p><p class="medium-feed-snippet">Databricks provides some pretty good functionality out the box, but as dashboard designers, sometimes we would like to add our own styling&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@psrossman/databricks-dashboards-building-kpi-indicators-with-html-styling-78a9275d7f5d?source=rss-6903991ebf07------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@psrossman/databricks-dashboards-building-kpi-indicators-with-html-styling-78a9275d7f5d?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/78a9275d7f5d</guid>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[bi]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sun, 10 Aug 2025 20:33:24 GMT</pubDate>
            <atom:updated>2025-08-10T20:33:24.723Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[A Tutorial on the New Features released in Databricks Dashboards (July 2025)]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@psrossman/a-tutorial-on-the-new-features-released-in-databricks-dashboards-july-2025-75dbcc2ded4e?source=rss-6903991ebf07------2"><img src="https://cdn-images-1.medium.com/max/1704/1*1gffZOu1gdEGMVIOwQh81A.png" width="1704"></a></p><p class="medium-feed-snippet">This post recaps the new features released here for Databricks dashboards, and giving a quick tutorial of each.</p><p class="medium-feed-link"><a href="https://medium.com/@psrossman/a-tutorial-on-the-new-features-released-in-databricks-dashboards-july-2025-75dbcc2ded4e?source=rss-6903991ebf07------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@psrossman/a-tutorial-on-the-new-features-released-in-databricks-dashboards-july-2025-75dbcc2ded4e?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/75dbcc2ded4e</guid>
            <category><![CDATA[bi]]></category>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[dashboard]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sun, 03 Aug 2025 14:40:51 GMT</pubDate>
            <atom:updated>2025-08-03T14:40:51.888Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Extending Databricks Dashboards with HTML]]></title>
            <link>https://medium.com/@psrossman/extending-databricks-dashboards-with-html-c5db569376f6?source=rss-6903991ebf07------2</link>
            <guid isPermaLink="false">https://medium.com/p/c5db569376f6</guid>
            <category><![CDATA[databricks]]></category>
            <category><![CDATA[dashboard]]></category>
            <category><![CDATA[databricks-basics]]></category>
            <dc:creator><![CDATA[Paul Rossman]]></dc:creator>
            <pubDate>Sun, 27 Jul 2025 13:15:12 GMT</pubDate>
            <atom:updated>2025-07-27T13:15:12.852Z</atom:updated>
            <content:encoded><![CDATA[<p>Hello all — Welcome to my first post on Medium. In this post, I would like to explain how to use some simple HTML code to give your Databricks dashboards some extra style compared to what comes out of the box.</p><p><strong>Getting Started</strong></p><p>To get started, if you haven’t signed up for the free version of Databricks, click <a href="https://www.databricks.com/learn/free-edition">here</a>. All you need is an email, and you are up and running.</p><p>On the left hand navigation, click dashboards and new dashboard. Next, click on Add data, and using the catalog, navigate to <strong>samples.bakehouse.sales_transactions. </strong>It will produce something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u0rjiVfAjm-RVYqF_i-agA.png" /></figure><p>Now, we can do a lot out of the box with this data. Creating charts, tables, but we can also use HTML to create some KPI indicators.</p><p><strong>Compare Day over Day Sales</strong></p><p>Let’s say we want to compare day over day sales for our bakehouse. First, let’s construct a query, using a lag function to pull the current day and previous day’s sales.</p><pre>WITH aggregated_sales AS (<br>  SELECT <br>    to_date(DATE_TRUNC(&#39;day&#39;, dateTime)) as start_date,<br>    SUM(totalPrice) as totalSales<br>  FROM samples.bakehouse.sales_transactions<br>  GROUP BY DATE_TRUNC(&#39;day&#39;, dateTime)<br>)<br>SELECT <br>  start_date,<br>  totalSales,<br>  LAG(totalSales) OVER (ORDER BY start_date) as previousDaySales<br>FROM aggregated_sales<br>ORDER BY start_date</pre><p>Next, let’s calculate the absolute change and percent variance over the prior day.</p><pre>WITH aggregated_sales AS (<br>  SELECT <br>    to_date(DATE_TRUNC(&#39;day&#39;, dateTime)) as start_date,<br>    SUM(totalPrice) as totalSales<br>  FROM samples.bakehouse.sales_transactions<br>  GROUP BY DATE_TRUNC(&#39;day&#39;, dateTime)<br>)<br>SELECT <br>  start_date,<br>  totalSales,<br>  LAG(totalSales) OVER (ORDER BY start_date) as previousDaySales,<br>  totalSales-previousDaySales as abs_change,<br>  abs_change/previousDaySales as percent_change<br>FROM aggregated_sales<br>ORDER BY start_date</pre><p>Next, let’s add an indicator using a case statement and unicode. My favorite site for finding unicode values is <a href="https://www.compart.com/en/unicode/">https://www.compart.com/en/unicode/</a></p><pre>WITH aggregated_sales AS (<br>  SELECT <br>    to_date(DATE_TRUNC(&#39;day&#39;, dateTime)) as start_date,<br>    SUM(totalPrice) as totalSales<br>  FROM samples.bakehouse.sales_transactions<br>  GROUP BY DATE_TRUNC(&#39;day&#39;, dateTime)<br>)<br>SELECT <br>  start_date,<br>  totalSales,<br>  LAG(totalSales) OVER (ORDER BY start_date) as previousDaySales,<br>  totalSales-previousDaySales as abs_change,<br>  abs_change/previousDaySales as percent_change,<br>  case when abs_change &gt;= 0 then &#39;&amp;#9989&#39; when abs_change &lt; 0 then &#39;&amp;#10060&#39; else &#39;&amp;#128309&#39; end as indicator<br>FROM aggregated_sales<br>ORDER BY start_date</pre><p>Finally, let’s add it to our dashboard, in a simple table form.</p><p>First, add a chart from the menu below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/208/1*zBcvI1dI2Zdx1VSbLOGztQ.png" /></figure><p>On the right side panel, navigate to table, then add columns (uncheck the eye with the diagonal line) where necessary. When it comes to the indicator and showing it as a symbol instead of text, click on the “Allow HTML content” and it will show as the unicode symbol.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/247/1*O2rUVPu_T_DkHeqSBVq5kQ.png" /></figure><p>With a little extra number formatting, and renaming of columns, we can produce something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/877/1*dW_twT8n0fmvtLIY-jtzkA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/840/1*zZe_wrkQ63sU9wKNK2vxfw.png" /></figure><p>In an upcoming post, I will show how to use HTML tables and CSS to create along with the unicode values to produce professional looking KPI indicators.</p><p>Thank you for reading!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c5db569376f6" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>