<?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 Yubi on Medium]]></title>
        <description><![CDATA[Stories by Yubi on Medium]]></description>
        <link>https://medium.com/@credavenue?source=rss-2c46741ffea6------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*dKOb82vqIn09x0N_.</url>
            <title>Stories by Yubi on Medium</title>
            <link>https://medium.com/@credavenue?source=rss-2c46741ffea6------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 09:45:59 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@credavenue/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[How Digital Platform with Advanced Artificial Intelligence Technology are Customizing User…]]></title>
            <link>https://medium.com/vivriticapital/how-digital-platform-with-advanced-artificial-intelligence-technology-are-customizing-user-457207bc747?source=rss-2c46741ffea6------2</link>
            <guid isPermaLink="false">https://medium.com/p/457207bc747</guid>
            <category><![CDATA[financial-services]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[fintech]]></category>
            <dc:creator><![CDATA[Yubi]]></dc:creator>
            <pubDate>Fri, 16 Nov 2018 08:47:11 GMT</pubDate>
            <atom:updated>2018-11-17T05:32:34.192Z</atom:updated>
            <content:encoded><![CDATA[<h3>How Digital Platform with Advanced Artificial Intelligence Technology are Customizing User Experience — by Gaurav Kumar</h3><h4><strong>The combination of AI technologies such as machine learning and NLP along with data science and analytics have emerged as important tools for companies to streamline their personalisation activities such as in digital marketing and recommendation engines.</strong></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2q73DqQyNTTCwQ4k8frlww.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/photos/vJQCyAOegiI?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">yatharth roy vibhakar</a> on <a href="https://unsplash.com/search/photos/artificial-intelligence?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>I n the not so distant past, a company could easily gain the loyalty of its customers by simply selling superior products or services. However, as the market saw a deluge of numerous similar products and services with little to set them apart in terms of value, companies had to constantly be looking for innovative ways to acquire and retain customers. As brands realised the role of customer or user experience (UX) in a person’s lifecycle, they began to seek ways in which to deliver additional unique value across all stages.</p><p>As consumers started interacting and transacting with brands through online or digital channels, companies leveraged this trend by giving customers the option of personalising products and services to meet their requirements. Today, most companies across various industries such as e-commerce, travel, education, etc. will concur on the importance of personalisation in winning customers and building brand loyalty among them. And truly, personalisation has become an important tool and a key element of any brand’s value proposition, which if delivered effectively, can bring it great success. The effectiveness of the tool is particularly notable on digital or mobile platforms, thanks to AI techniques such as machine learning and natural language processing which are enabling brands to use customer insights in ways one could never imagine before.</p><p><strong>How machine learning and automation helps personalise the customer experience and journey</strong></p><p>In the financial services industry too, the use of Artificial intelligence has evolved considerably over the past couple of years thanks to the increasing popularity and adoption of digital platforms and the push towards digitisation, presenting a significant opportunity for the service providers to reach out to consumers directly over online channels. Also, such channels are a rich source of data that can be leveraged to deliver greater personalisation to customers and ensure higher levels of satisfaction. The combination of AI technologies such as machine learning and NLP along with data science and analytics have emerged as important tools for companies to streamline their personalisation activities such as in digital marketing and recommendation engines. More than anything, big data and AI allows companies to make sense of large volumes of data from disparate sources and apply algorithms on such data sets to derive useful and unique insights on customers, find out about their specific needs and wants, their propensity to continue using their services, and loyalty to their brand. This allows companies to tailor their offerings to deliver additional value as compared to their competitors. Some of the major areas where artificial intelligence has had the most impact is the creation of chatbots (Natural Language Processing) and development of new-age credit scoring models (machine learning). Let’s look how.</p><p><strong>Credit scoring models</strong></p><p>Credit scoring models employ a host of supervised machine learning algorithms to accelerate lending decisions and make them more data-driven, while potentially limiting future risk. Customer’s historical financial transaction and payment history data from financial institutions serve as the foundation of most credit scoring models which use tools such as regression, decision trees, and statistical analysis to generate a credit score using limited amounts of structured data.</p><p>Alternative credit scoring models, on the other hand, use a range of one-dimensional and multidimensional data available in the digital footprints left by users on websites like Facebook, Twitter, and LinkedIn. The information provided by these data sources goes beyond the usual financial and credit history of EMIs and credit card payments, and instead looks at different aspects of an individual’s personality to determine their creditworthiness.</p><p>Some of the commonly used alternate date techniques leverage upon bank account history of customer. The bank statements are used as source to analyze spending patterns, constant payment patterns of existing EMIs, average balance maintained etc. Another alternative to estimate customer income could be tie-ups with telecom companies to leverage data on customers like number of instances a prepaid customer’s mobile balance was zero, time taken to recharge the phone once the balance was zero on prepaid customer and parameters like international calls, roaming, call history and consistency of balance.</p><p>The real challenge lies in mining the available data effectively and making meaningful and sensible correlations using algorithms and analysis. Sensible and accurate correlations can help predict a person’s credit behaviour and introduce potentially underserved customers to the formal credit market.</p><p>Using traditional and alternate sources of data available, financial institutions perform feature engineering on the dataset; identifying statistically significant variables which enables scoring models to differentiate between good and bad borrowers.</p><p>The key going forward will be track the effectiveness of these credit scoring models. While lending based on these models is happening big time and it is bringing in efficiency, effectiveness and credibility of the same is unknown and largely untested.</p><p><strong>Product Recommendation System</strong></p><p>Financial institutions offer various services to customers ranging from different products like fixed deposits, credit cards, loans, insurance etc. The objective of recommendation systems is to find out the merchants/items which a customer might be interested in after already availing certain services from bank. Financial institutions have started taking a much broader view on customers beyond one transaction and look at metric like Customer Lifetime Profitability Value. Customer lifetime value (or CLV) is the worth of a customer over the length of their entire tenure. Using customer lifetime value as a metric requires your entire bank to shift its focus from quarterly profits to the long-term health of customer relationships.</p><p>CLV also provides an upper limit on customer acquisition costs and provides insights into customer engagement by making informed decisions on product &amp; pricing. It also differentiates between customers which are worth pursuing for long-term value and which ones are likely to switch over for balance transfer. Understanding the difference between different types of customers guides enables these firms to maximize their value and cater to different segments appropriately. It estimates the profit &amp; loss if many competing items can be recommended to the customer. Now based on the profile of the customer, recommends a customer centric or product centric offering. High value customers, which other institutions are also interested to gain wallet share, are usually provided best of your offers to improve customer loyalty. Recommendation system are used to finds out the products which a customer might be interested into after buying something else. It estimates the profit &amp; loss if many competing products can be recommended to the customer.</p><p><strong>Risk Management &amp; Fraud Detection</strong></p><p>Identifying individuals who have a high risk for fraud is a task inherently suitable for machine learning. These machine learning solutions can comb through huge transactional datasets and identify all cases that might be prone to fraud. While providing a loan to any client; financial institutions go through a process of risk assessment to estimate the creditworthiness of a prospect. Traditional systems relied on historical data like transaction history, credit history and income growth over years to understand the risk associated with every loan extended. This resulted in inconsistent estimates as historical data is not always an accurate standard to predict future behaviour.</p><p>With the help of machine learning, fraud detection techniques can be made efficient and effective. The solutions created can analyze historical transaction data to build a model that can detect fraudulent patterns. Companies are also using machine learning for biometric authentication by collecting data from various sources and then mapping them to trigger points, AI solutions can find out the rate of defaulting or fraudulence for each potential customer thus alerting the financial institutions beforehand that giving any credit to these individuals is risky.</p><p><em>Originally published at </em><a href="http://bwdisrupt.businessworld.in/article/How-Digital-Platform-with-Advanced-Artificial-Intelligence-Technology-are-Customizing-User-Experience/18-09-2018-160251/"><em>bwdisrupt.businessworld.in</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=457207bc747" width="1" height="1" alt=""><hr><p><a href="https://medium.com/vivriticapital/how-digital-platform-with-advanced-artificial-intelligence-technology-are-customizing-user-457207bc747">How Digital Platform with Advanced Artificial Intelligence Technology are Customizing User…</a> was originally published in <a href="https://medium.com/vivriticapital">Vivriti Capital</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mediawiki and Azure Single Sign On]]></title>
            <link>https://medium.com/vivritiengineering/mediawiki-and-azure-single-sign-on-e3fbc13d1f46?source=rss-2c46741ffea6------2</link>
            <guid isPermaLink="false">https://medium.com/p/e3fbc13d1f46</guid>
            <category><![CDATA[aml]]></category>
            <category><![CDATA[single-sign-on]]></category>
            <category><![CDATA[azure-active-directory]]></category>
            <category><![CDATA[wiki]]></category>
            <category><![CDATA[azure-ad]]></category>
            <dc:creator><![CDATA[Yubi]]></dc:creator>
            <pubDate>Tue, 16 Oct 2018 15:29:55 GMT</pubDate>
            <atom:updated>2018-10-16T15:29:55.360Z</atom:updated>
            <content:encoded><![CDATA[<h3>Mediawiki and Azure AD Single Sign On</h3><p>W e wanted to build our Intranet solution and wiki seemed like a good choice. We chose to use mediawiki, the most popular wiki engine. We set it up using a standard Bitnami installation found on AWS marketplace. At the time we set this up, we were on version 1.31 of mediawiki. To ensure ease of use of the same, we also set up single sign on integration for the same. Find our journey to setting up single sign on with Azure AD. This post assumes a standard bitnami mediawiki over apache2 on a Ubuntu 16.04 machine. But the links provided in this post should help for other variations.</p><h3>Mediawiki Setup (High Level Set up)</h3><ol><li>If you do not have an AWS account, sign up for on</li><li>Log into the AWS Console</li><li>In the same browser window, search for aws marketplace</li><li>Search for <strong><em>bitnami mediawiki</em></strong> subscribe to the software, choose your machine and aws region and launch via web.</li><li>If you create new secure keys, do download the pem file to be used in putty to securely log into the machine</li><li>Do install into the right VPC, subnets etc.</li><li>You can track the initiation of the machine from the EC2 portal.</li><li>Go to the EC2 portal and get the username (usually <strong><em>user</em></strong>) and password. See <a href="https://docs.bitnami.com/aws/faq/get-started/find-credentials/">How to get bitnami credentials?</a> for more information</li><li>Log into the Ubuntu machine. See instructions at <a href="https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/putty.html">Connect to your Linux instance</a> for more information. User name will be <strong><em>bitnami</em> </strong>(See <a href="https://docs.bitnami.com/aws/faq/get-started/connect-ssh/">Connect to Bitnami Virtual machine</a> for more information)</li></ol><h3>Generating a Certificate</h3><ol><li>Log into the machine using putty (from a Windows machine) or another mechanism</li><li>Here will use a the letsencrypt script for generating a certificate. (You can use other mechanisms to generate the certificate or buy one)</li><li>Use the following set of commands to generate certificates</li></ol><pre>cd /tmp<br>git clone <a href="https://github.com/letsencrypt/letsencrypt">https://github.com/letsencrypt/letsencrypt</a><br>cd letsencrypt<br>./letsencrypt-auto<br>cd /tmp/letsencrypt/<br>./letsencrypt-auto certonly --webroot -w /opt/bitnami/apps/mediawiki/htdocs -d int.example.com</pre><p>4. The above commands will generate the certificates and place them in the directory /etc/letsencrypt.</p><p>5. You will see that certificates have been generated in the folder <strong>/etc/letsencrypt/live/int.example.com</strong>. (You might need to have root access to be able to cd into and view this)</p><p>6. <strong>References: </strong><a href="https://community.bitnami.com/t/let-s-encrypt-on-bitnami-mediawiki-image/45038/3">https://community.bitnami.com/t/let-s-encrypt-on-bitnami-mediawiki-image/45038/3</a></p><p>7. Note: You may need to have all your ports opened for this to work correctly. In AWS, you may want to navigate to the security group and allow all traffic (in and out). See AWS documentation for opening ports</p><h3>Installing Certificate on the Apache Server</h3><ol><li>Copy the certificates to the Apache config folders with the following commands</li></ol><pre>cp /etc/letsencrypt/live/int.example.com/fullchain.pem /opt/bitnami/apache2/conf/server.crt<br>cp /etc/letsencrypt/live/int.example.com/privkey.pem /opt/bitnami/apache2/conf/server.key</pre><p>2. Restart your apache server using the following command</p><pre>sudo /opt/bitnami/ctlscript.sh restart apache</pre><p>3. Now you should be able to access your mediawiki installation at <a href="https://int.example.com./">https://int.example.com.</a></p><h3>Redirecting from port 80 to 443</h3><ol><li>Users typically access <a href="http://int.example.com/">http://int.example.com</a> but this is an insecure port. Hence, you will need to redirect all users from port 80 to port 443.</li><li>You will need to modify the file <strong><em>bitnami.conf</em></strong><em> </em>to redirect traffic from port 80 to port 443. See instructions at <a href="https://docs.bitnami.com/aws/apps/mediawiki/administration/force-https-apache/">Force http to https</a>.</li><li>Use the following commands (with vi as editor)</li></ol><pre>cd /opt/bitnami/apache2/conf/bitnami<br>vi bitnami.conf<br>... Modify file as in step #4 ...<br>sudo /opt/bitnami/ctlscript.sh restart apache</pre><p>4. Add the following lines under the VirtualHost directive for port 80</p><pre>  RewriteEngine On<br>  RewriteCond %{HTTPS} !=on<br>  RewriteRule ^/(.*) https://%{SERVER_NAME}/$1 [R,L</pre><p>5. Restart the apache server using the command as in #3</p><p><strong>Installing SimpleSamlphp</strong></p><ol><li>Now, we can install <a href="https://simplesamlphp.org/download">simplesamlphp</a></li><li>You can get the download link by click on the archive link and getting the URL for the latest version. Use the following commands to download and install the simplesamlphp library. See <a href="https://simplesamlphp.org/docs/stable/simplesamlphp-install#section_4">link</a></li></ol><pre>cd /tmp<br>wget <a href="https://github.com/simplesamlphp/simplesamlphp/releases/download/v1.16.2/simplesamlphp-1.16.2.tar.gz">https://github.com/simplesamlphp/simplesamlphp/releases/download/v1.16.2/simplesamlphp-1.16.2.tar.gz</a><br>tar xzf simplesamlphp-1.x.y.tar.gz<br>sudo mv simplesamlphp-1.16.2 /etc/simplesamlphp</pre><p>3. Navigate to your document root at /opt/bitnami/apps/mediawiki/htdocs</p><p>4. Add a soft link to simplesaml application with the following command</p><pre>ln -sf /etc/simplesamlphp/www simplesaml</pre><p>5. Now, you should be able to access the simlpesaml application with the link <a href="https://intranet.vivriticapital.com/simplesaml">https://int.example.com/simplesaml</a></p><p>6. Modify the config file to change the password for simplesaml tool portal</p><pre>cd /etc/simplesamlphp/config<br>vi config.php</pre><p>7. Find the line <strong><em>‘auth.adminpassword’ =&gt; ‘123’,</em></strong></p><p>8. Change the password from 123 to say <strong><em>password</em></strong> or some other password. Best to choose a password that nobody can guess</p><p>9. Restart your apache server with the following command</p><pre>sudo /opt/bitnami/ctlscript.sh restart apache</pre><h4>Importing Metadata into SimpleSaml</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RruJDaJ_MjIf6aToCEsDXg.png" /><figcaption>FederationData.xml</figcaption></figure><ol><li>When users access your page, they will need to be redirected to Azure for authentication.</li><li>To permit your application to access your Azure AD sign on page, you will need to provide some metadata for simplesaml.</li><li>Head over to your <a href="https://portal.azure.com/">Azure portal</a>. Login as an administrator with the right permissions</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9eChzOC2k07ihFiudZVBFQ.png" /></figure><p>4. Click on Azure Active Directory -&gt; App Registrations -&gt; Endpoints</p><p>5. Copy the URL where you can access your FederationMedata.xml file.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jwh8-Ny8eUqyMZfZAoYOjg.png" /></figure><p>6. Paste the URL in a new window to get the FederationMetadata.xml file</p><p>7. Right click and save the XML file as FederationMetadata.xml file.</p><p>8. Now head back to the simple saml tool at <a href="https://intranet.vivriticapital.com/simplesaml">http://int.example.com/simplesaml</a></p><p>9. Choose the Federation tab and then click the <strong><em>XML to SimpleSAMLphp metadata converter </em></strong>link</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PbGPbgs_m8tdLurQYo2jsw.png" /></figure><p>10. Paste the XML contents of the <strong><em>FederationMetadata.xml</em></strong> file</p><p>11. Click Parse to convert it to simplesamlphp format</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fOMrdUITQGsi7Ulc5x2FuA.png" /></figure><p>12. Copy the contents of the parse data in the window named saml20-idp-remote</p><p>13. Head back to your putty session. Use the following commands to store the metadata for simplesaml to access</p><pre>cd /etc/simplesamlphp/metadata<br>vi saml20-idp-remote.php</pre><p>14. Paste the contents (metadata in php format) and save the file</p><p>15. For future use note down the link in the first line of the metatata — eg. <a href="https://sts.windows.net/54526d45-a91c-455a-a023-49951eceaaa5/">https://sts.windows.net/54526d45-a91c-432a-a023-49951efeadb5/</a></p><p><strong>Azure AD Single Sign On Set up</strong></p><ol><li>Next, head over to the azure portal (portal.azure.com)</li><li>On the left side click on Azure Active Directory. Select App Registrations and View All Applications.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WPHKsCuOgmmQgcb1O-AQKw.png" /><figcaption>Azure Portal App Set up</figcaption></figure><p>3. Select New Application Registration</p><p>4. For Name provide a name of your choice — eg. Example.com wiki</p><p>5. The home page URL for your wiki (Note that you should have, by now, set up a CNAME to redirect the URL to your AWS EC2 host)</p><p>6. Click Create at the bottom to create your wiki. It will lead you to the page as shown below</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8LoWCpKv5NLo1x6-3A4XUQ.png" /></figure><p>7. Click on Settings. Keep this page/tab open since we will need to come back to this</p><p>8. Note down the GUID that is the Application Id</p><h4>Setting up simplesamlphp configuration</h4><ol><li>Now head back to your putty session. Use the following commands to set up the configuration for simplesamlphp</li></ol><pre>cd /etc/simplesamlphp/config<br>vi authsources.php</pre><p>2. Find the section for <strong><em>default-sp</em></strong> (‘default-sp’ =&gt; array)</p><p>3. Set the value of entityId as the value of the application id from Azure portal. Prefix this with the string spn. Eg. — <strong><em>spn:7a5976f1-bbb0–40ef-9d85-d80b3b70bac7</em></strong></p><p>4. Set the value of idp to the value we noted down from the converted federationmedata.xml file — eg. <a href="https://sts.windows.net/54526d45-a91c-455a-a023-49951eceaaa5/"><strong><em>https://sts.windows.net/54526d45-a91c-432a-a023-49951efeadb5/</em></strong></a></p><p>5. Leave the rest of the parameters as is. Save the file and exit the editor</p><p>6. Restart your apache server as done before</p><p>7. Head over to the Azure portal. Go to App Registrations. Then click on the View All Applications and choose your application <strong><em>Example.com wiki</em></strong><em>. </em>Click <strong><em>Settings</em></strong> followed by <strong><em>Reply URLs</em></strong><em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pOzXGmpuo9UsNTGPLqjEkA.png" /></figure><p>8. Remove the existing value for the ReplyURL and replace with <a href="https://int.example.com/simplesaml/module.php/saml/sp/saml2-acs.php/default-sp"><strong><em>https://int.example.com/simplesaml/module.php/saml/sp/saml2-acs.php/default-sp</em></strong></a><strong><em> </em></strong>as shown below</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iATJio0ITnZTa0AnXmlFkw.png" /></figure><p>9. Now access your simplesaml web link at <a href="https://intranet.vivriticapital.com/simplesaml">https://intranet.vivriticapital.com/simplesaml</a></p><p>8. Navigate to the Authentication tab. Click <strong><em>Test Authentication Sources</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Vxubz7qb9zHbVIIZcRDocQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/963/1*PU3EavcRQ0-uk7UdwY62zA.png" /></figure><p>9. Click <strong><em>default-sp</em></strong></p><p>10. This should lead you to the Azure AD/Office 365 sign in page. Login with your credentials</p><p>11. This should lead to a page as shown below. These are the values returned by the Identity Provider, MicroSoft, back to simplesampl and consequently your application</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cmFjzdiGG5nBoiMKyOM45g.png" /><figcaption><strong><em>Simple SAML 2.0 Demo Example</em></strong></figcaption></figure><p>12. Do take some time to go through these attributes since you will be using some of these when you set up the single sign on for your mediawiki site.</p><p><strong>Install PluggableAuth and SimpleSaml mediawiki extensions</strong></p><ol><li><a href="https://www.mediawiki.org/wiki/Extension:PluggableAuth">PluggableAuth</a> is an extension for mediawiki to use simplesaml based single sign on.</li><li>Follow the instructions in the above page to download and install PluggableAuth and configure it. You can use these instructions to configure PluggableAuth. You can leave out <strong><em>$wgPluggableAuth_Class</em></strong> set up for now. We will deal with it in the next step.</li><li>Next, download and install the <a href="https://www.mediawiki.org/wiki/Extension:SimpleSAMLphp">SimpleSaml mediawiki</a> extension as outlined in the link.</li><li>Configure both the plugins as explained in the next few steps.</li><li>First configure mediawiki options. Instructions and information can be found on the <a href="https://www.mediawiki.org/wiki/Manual:User_rights#List_of_permissions">User Rights</a> wiki</li></ol><pre>$wgGroupPermissions[&#39;*&#39;][&#39;edit&#39;] = false;<br>$wgGroupPermissions[&#39;user&#39;][&#39;edit&#39;] = true;</pre><pre>$wgGroupPermissions[&#39;*&#39;][&#39;createaccount&#39;] = true;<br>$wgGroupPermissions[&#39;*&#39;][&#39;autocreateaccount&#39;] = true;</pre><p>6. The first line forces users to log in before they can edit or contribute to page. If they are not logged in, then they will only be allowed to read and not contribute</p><p>7. The second line allows all users to edit though this line isnt really needed</p><p>8. The next two lines allows autocreation of accounts. So, once Azure AD authenticates users, an account will automatically be created if it doesnt exist for a given user.</p><p>9. Next, we will configure PluggableAuth</p><pre>wfLoadExtension( &#39;PluggableAuth&#39; );</pre><pre>$wgPluggableAuth_EnableAutoLogin = false;<br>$wgPluggableAuth_EnableLocalLogin = false;<br>$wgPluggableAuth_EnableLocalProperties = true;<br>$wgPluggableAuth_ButtonLabelMessage = &#39;Login&#39;;<br>$wgPluggableAuth_Class = &#39;SimpleSAMLphp&#39;;</pre><p>10. First line adds the extension PluggableAuth.</p><p>11. If EnableAutoLogin is true, then users will automatically be logged in. Else, they will need to click on the login link and log in. So, it allows users to be in either read only or edit mode.</p><p>12. EnableLocalLogin provides users the option of logging in directly using the mediawiki user table. This may needed for the primary administrator login. So, I left it as is. There are other ways of achieving this too.</p><p>13. ButtonLabelMesage — an optional string for the label</p><p>14. $wgPluggableAuth_Class — Set this to SimpleSAMLphp for now. We will configure SimpleSAMLphp in the next few steps.</p><pre>$wfLoadExtension( &#39;SimpleSAMLphp&#39; );</pre><pre>$wgSimpleSAMLphp_InstallDir = &#39;/etc/simplesamlphp&#39;;<br>$wgSimpleSAMLphp_AuthSourceId = &#39;default-sp&#39;;</pre><pre>$wgSimpleSAMLphp_UsernameAttribute = [&#39;<a href="http://schemas.xmlsoap.org/ws/2005/05/identity/claims/givenname&#39;,&#39;http://schemas.xmlsoap.org/ws/2005/05/identity/claims/surname&#39;">http://schemas.xmlsoap.org/ws/2005/05/identity/claims/givenname&#39;,&#39;http://schemas.xmlsoap.org/ws/2005/05/identity/claims/surname&#39;</a>];<br>$wgSimpleSAMLphp_EmailAttribute = &#39;<a href="http://schemas.xmlsoap.org/ws/2005/05/identity/claims/name&#39;">http://schemas.xmlsoap.org/ws/2005/05/identity/claims/name&#39;</a>;<br>$wgSimpleSAMLphp_RealNameAttribute = [<br>        &#39;<a href="http://schemas.xmlsoap.org/ws/2005/05/identity/claims/givenname&#39;">http://schemas.xmlsoap.org/ws/2005/05/identity/claims/givenname&#39;</a>,<br>        &#39;<a href="http://schemas.xmlsoap.org/ws/2005/05/identity/claims/surname&#39;">http://schemas.xmlsoap.org/ws/2005/05/identity/claims/surname&#39;</a><br>];</pre><p>15. First line adds the extension SimpleSAMLphp mediawiki extension</p><p>16. Second line provides the path to the SimpleSAMLphp installation. In our case, we have installed it in /etc/simplesamlphp</p><p>17. $wgSimpleSAMLphp_AuthSourceId provides the configuration group key under which we have added the authentication source. In our case we used <strong><em>default-sp</em></strong></p><p>18. The next set of values defines the mapping between the attribute name-value pairs returned by Azure AD and mediawiki’s login parameters. You can refer back to the <strong><em>Simple SAML 2.0 Demo Example </em></strong>screen shot to see how this is set up.</p><p>19. $wgSimpleSAMLphp_UsernameAttribute — This is the name under which the wiki contributions will be displayed. So, it is best to keep these unique. Since Azure AD does not have a kerberos like unique name, we have used a combination of {firstname, lastname} aka {givenname, surname} for this field. The input is an array of these attribute values. I have also provided a modified version of the SimpleSAMLphp extension class that handles this. Code for the same is provided in the next section</p><p>20. Similarly, set up $wgSimpleSAMLphp_EmailAttribute and $wgSimpleSAMLphp_RealNameAttribute</p><p>21. Finally, we have changed the code a bit to suit Azure AD. Change directory to /opt/bitnami/apps/mediawiki/htdocs/extensions/SimpleSAMLphp. Open the file and find the following code snippet.</p><pre>if ( isset( $GLOBALS[&#39;wgSimpleSAMLphp_UsernameAttribute&#39;] ) ) {<br>                        if ( array_key_exists( $GLOBALS[&#39;wgSimpleSAMLphp_UsernameAttribute&#39;],<br>                                $attributes ) ) {<br>                                $username = strtolower(<br>                                        $attributes[$GLOBALS[&#39;wgSimpleSAMLphp_UsernameAttribute&#39;]][0] );<br>                                $nt = Title::makeTitleSafe( NS_USER, $username );<br>                                if ( is_null( $nt ) ) {<br>                                        return false;<br>                                }<br>                                $username = $nt-&gt;getText();<br>                                $id = User::idFromName( $username );<br>                        } else {<br>                                wfDebug( &#39;SimpleSAMLphp: Could not find username attribute &#39; .<br>                                        $attributes );<br>                                return false;<br>                        }<br>                } else {<br>                        wfDebug( &#39;SimpleSAMLphp: $wgSimpleSAMLphp_UsernameAttribute is not set&#39; );<br>                        return false;<br>                }</pre><p>22. Replace the above code snipped with the one below.</p><pre>if ( isset( $GLOBALS[&#39;wgSimpleSAMLphp_UsernameAttribute&#39;] ) ) {<br>                        $userNameAttribute = $GLOBALS[&#39;wgSimpleSAMLphp_UsernameAttribute&#39;];<br>                        if ( is_array( $userNameAttribute ) ) {<br>                                $username = &quot;&quot;;<br>                                foreach ( $userNameAttribute as $attribute ) {<br>                                        if ( array_key_exists( $attribute, $attributes ) ) {<br>                                                if ( $username != &quot;&quot; ) {<br>                                                        $username .= &quot; &quot;;<br>                                                }<br>                                                $username .= $attributes[$attribute][0];<br>                                        } else {<br>                                                wfDebug( &#39;SimpleSAMLphp: Could not find user name attribute &#39; .<br>                                                        $attribute );<br>                                                return false;<br>                                        }<br>                                }<br>                        } else {<br>                                if ( array_key_exists( $userNameAttribute, $attributes ) ) {<br>                                        $realname = $attributes[$userNameAttribute][0];<br>                                } else {<br>                                        wfDebug( &#39;SimpleSAMLphp: Could not find user name attribute &#39; .<br>                                                $attributes );<br>                                        return false;<br>                                }<br>                        }</pre><pre>} else {<br>                        wfDebug( &#39;SimpleSAMLphp: $wgSimpleSAMLphp_UsernameAttribute is not set&#39; );<br>                        return false;<br>                }</pre><p>23. Restart the apache server with the command <strong><em>sudo /opt/bitnami/ctlscript.sh restart apache</em></strong></p><h4>Handling a bug</h4><ol><li>See <a href="https://www.mediawiki.org/wiki/Extension:SimpleSAMLphp#Known_Bugs">Known Bugs</a> for more information on a bug that will cause this to not work.</li><li>To handle this, make the following changes.</li><li>Edit the LocalSettings.php file in the folder /opt/bitnami/apps/mediawiki/htdocs</li><li>Change <strong><em>$wgMainCacheType = CACHE_NONE; </em></strong>to <strong><em>$wgMainCacheType = CACHE_DB;</em></strong></li><li>Edit the simplesaml config file in /etc/simplesamlphp/config/config.php</li><li>Set the two values as shown below</li></ol><pre>&#39;store.type&#39;                    =&gt; &#39;sql&#39;,<br>&#39;store.sql.dsn&#39;                 =&gt; &#39;sqlite:/opt/bitnami/apache2/var/cache/sqlitedatabase.sq3&#39;,</pre><p>7. Restart the apache server with the command <strong><em>sudo /opt/bitnami/ctlscript.sh restart apache</em></strong></p><h4><strong>Testing and Usage</strong></h4><ol><li>You are all set now. You can now access your wiki at <a href="http://int.example.com./">http://int.example.com.</a></li><li>If you click on the Log In link on the right top corner, it should redirect you to your microsoft onlin login page. Once you login, you will be redirected back to the home page logged in with a user name</li><li>A user account is created in the backend automatically</li></ol><p><strong>Happy Wikiying — Bye!</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e3fbc13d1f46" width="1" height="1" alt=""><hr><p><a href="https://medium.com/vivritiengineering/mediawiki-and-azure-single-sign-on-e3fbc13d1f46">Mediawiki and Azure Single Sign On</a> was originally published in <a href="https://medium.com/vivritiengineering">VivritiEngineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Multiple Artifacts in AWS Codepipeline and CodeBuild]]></title>
            <link>https://medium.com/vivritiengineering/multiple-artifacts-in-aws-codepipeline-and-codebuild-9f2d22eb822a?source=rss-2c46741ffea6------2</link>
            <guid isPermaLink="false">https://medium.com/p/9f2d22eb822a</guid>
            <category><![CDATA[deployment-pipelines]]></category>
            <category><![CDATA[continuous-deployment]]></category>
            <category><![CDATA[build]]></category>
            <category><![CDATA[aws]]></category>
            <dc:creator><![CDATA[Yubi]]></dc:creator>
            <pubDate>Thu, 23 Aug 2018 15:32:42 GMT</pubDate>
            <atom:updated>2018-08-24T10:00:07.389Z</atom:updated>
            <content:encoded><![CDATA[<p>We wanted to automate our build and deploy process and move to CI/CD. Since we run our infrastructure on AWS, our natural choice was AWS CodePipeline and AWS CodeBuild. We started with a module, a web application archive used as a background job, run on Elastic Beanstalk. The application had more than one jar file as a dependency. Unfortunately, if you use AWS CodeBuild, AWS CodePipeline allows just one artifact as input to the next step, a serious limitation, in our opinion. We considered a few different ways to solve this problem</p><ul><li>Not using Code Build — Jenkins was an option but then again this required some amount of set up</li><li>Putting up a local private maven repository or artifactory — Again, this involved some amount of set up which we did not want to invest in</li><li>Storing the output jar files in S3 and using commands in the buildspec file to download the dependencies to the local repository before using them in subsequent steps in the pipeline</li></ul><p>We decided to use the third approach since it seemed to be the easiest of the above and secondly, we simply expect AWS CodeBuild to catch up and allow us to use multiple artifacts as input to subsequent steps soon. Nevertheless, we had to cross several hurdles to get it working. This blog outlines the process we followed.</p><h3><strong>Setting up the Pipeline</strong></h3><h4>Source</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rA37T5mShq5CRuuYo9IXsw.png" /></figure><ol><li>Log into your AWS account</li><li>Choose the AWS Code Pipeline service</li><li>Start creating a new pipeline by clicking on Get Started and or Create New Pipeline</li><li>Provide a name for your pipeline — ours was called <strong><em>vivsecservice </em></strong>and click Next</li><li>Choose your source provider. We initially used AWS Codecommit but we could never get to making cross account access work and hence we moved our code to github. This blog explains the set up using github but this can as easily be set up using AWS CodeCommit</li><li>Once you choose GitHub, you will be asked to connect to github. Click, <strong><em>Connect to Gihub, </em></strong>provide your login details and connect to github. You will have a choice of repositories you can choose. You an choose the repository that contains the source code that will be built into the .war file. Choose the branch you want to build from. In our case, We choose repository called <strong><em>vivsecservice</em></strong> and the branch we are using is <strong><em>master</em></strong> since this setup is for a production build. Click <strong><em>Next</em></strong><em> </em>to proceed to the next step.</li></ol><h4>Build</h4><p>The next step is to set up the build process.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/791/1*yqh7zH8rNXYnmu9bXYcOQw.png" /></figure><ol><li>Choose AWS CodeBuild as the build provider (the whole purpose of this blog)</li><li>Choose the create a new build project option (Important)</li><li>Provide a name. We will use <strong><em>vivsecservice</em></strong></li><li>Set up the build environment. We will use Java8 on an Ubuntu machine</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/579/1*bKsCN4i2fdadKNzgMw5U0g.png" /></figure><p>5. Choose the No Cache option — This can be set up later if needed</p><p>6. Choose the <strong><em>Create a service role in your account</em></strong>. You can leave the pre-populated Role name as is. But do note this down since you will need this later. In our case the new role that will be created is <strong><em>code-build-vivsecservice-service-role</em></strong></p><p>7. Save the build project. Once the project is saved, you can click the <strong><em>View Project Details</em></strong> link.</p><p>8. This will lead you to the AWS Code Build Project page. Click on <strong><em>Edit Project</em></strong>. Go the <strong><em>Environment: How to build </em></strong>section and click on <strong><em>Update build specification</em></strong>. Simply enter the build specification file name. In our case, the build specification file name is <strong><em>buildspec-prod.yml</em></strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/744/1*OAIXGIJh1D0dDREz5d86nA.png" /></figure><p>9. Leave the rest of the data as is. Scroll down to the bottom and click <strong><em>Update</em></strong></p><p>10. Close the AWS CodeBuild window. Go back to the AWS Codepipeline window and click <strong><em>Next</em></strong></p><h4>Deploy</h4><ol><li>The next screen is the Deploy screen. We will deploy to AWS Elastic Beanstalk.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1021/1*oURGOM4s7QjL43dOqsx3Fg.png" /></figure><p>2. Provide the application and environment you would like to deploy the output war file to. In our case we are using an application called <strong><em>securitization-cashflowgen</em></strong> and the environment is <strong><em>securitization-cashflowgen-prod1</em></strong></p><p>3. Click <strong><em>Next</em></strong>. Click the <strong><em>Create role</em></strong> button. Use the default provided options. Click on the <strong><em>Allow</em></strong> button to create a new role that will be used. Going forward, for new pipelines, you can reuse this role. In our case a new role called <strong><em>AWS-CodePipeline-Service</em></strong> is created.</p><p>4. Click Next Step to proceed to the Review page. Review the details and click the <strong><em>Create Pipeline </em></strong>button to create the pipeline. You would have now created the pipeline. You would have now landed on the pipeline page for your pipeline.</p><h3>Editing the Pipeline to suit your needs</h3><h4>Source</h4><ol><li>As outlined before, we have multiple jars that will be built as part of this process and our eventual web application (<strong><em>vivsecservice</em></strong>) depends on these jars. So, now, we will modify the pipeline to address this case.</li><li>We will take a small subset of the problem to illustrate how to achieve what we set out to do. We have two libraries that we need to build, <strong><em>vivcommon </em></strong>and <strong><em>awsutils</em></strong>. <strong><em>awsutils </em></strong>build depends on the output of <strong><em>vivcommon</em></strong> build namely, <strong><em>vivcommon-1.0.0.jar</em></strong>.</li><li>First, we set up the build for vivcommon. Click on the <strong><em>Edit </em></strong>button to edit the pipeline. Next, click on the <strong><em>Edit</em></strong><em> </em>button to edit the stage. Remove the existing Source box (since we can’t rename the stage and we want to give it a good name, we are going to delete this action entirely and rebuild) by clicking on the <strong><em>x</em></strong>. Click on <strong><em>+</em></strong> to create a new action.</li><li>Choose <strong><em>Source </em></strong>for action category and fill up the rest of the details. We will call our Action name <strong><em>vivcommonsrc</em></strong>, Source Provider will be GitHub, repository <strong><em>vivriticapital/vivcommon</em></strong> and branch <strong><em>master</em></strong><em>. </em>(Note that you will need to click on the <strong><em>Connect to GitHub</em></strong> button to connect to GitHub). Scroll down to the bottom and in the provide a name for the output artifact #1. We will call our output artifact <strong><em>vivcommonsrc</em></strong><em>.</em></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/987/1*jKQLgzKQ74Zf1ljFOv2r9A.png" /></figure><p>5. Similarly, add the other source actions you will need. In our case, we will just add <strong><em>awsutils </em></strong>source action.</p><h4>Build</h4><ol><li>Next, we will set up the build step for <strong><em>vivcommon</em></strong>. Edit the build stage and remove the originally created action named <strong><em>CodeBuild</em></strong>. Click on <strong><em>+</em></strong> button to add a new action.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/986/1*bVvo_81KNtDXB5g89AWa0w.png" /></figure><p>2. Set Action Category to <strong><em>Build. </em></strong>Choose AWS CodeBuild as Build Provider. Action name is <strong><em>vivcommonbuild</em></strong>. Under Configure your project, choose <strong><em>Create a new build project. </em></strong>You can provide a project name of your choice. We will use <strong><em>vivcommonbuild. </em></strong>Set the environment details as necessary. We will set it as Ubuntu, Java and JDK8. Choose the <strong><em>Create a service role in your account</em></strong>. Leave the role name as the default value. The input artifact needs to be the output artifact of the Source step. Fill in details of output artificat. We will use <strong><em>vivcommonsrc</em></strong> and <strong><em>vivcommonout.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/990/1*kUc8V-78oKU4-7j1EhOe-Q.png" /></figure><p>3. Next, save the Build project. A build project is created and a link to the new build project will be provided. Click on the <strong><em>View project details</em></strong> link. On the AWS CodeBuild page, click on the <strong><em>Edit Project </em></strong>button.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/751/1*g1gWaYqtY3TnaBHu3RYAKA.png" /></figure><p>4. On this page, under the <strong><em>Environment: How to build </em></strong>section, click the <strong><em>Update build specification</em></strong> link. Enter a buildspec file name. Here we will use a custom buildspec file called <strong><em>buildspec-prod.yml</em></strong>. The contents of the build spec file are provided below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/945/1*4LEzHpqpOFtFKTeo6aPIwQ.png" /></figure><p>5. Essentially, in the post build command, we will upload the generated jar file to S3. So, this uploaded jar file plays the role of being the second (or nth) artifact of subsequent steps in the pipeline.</p><p>6. As can be seen above, the s3 destination bucket is provided. Since we will have different bucket names for different environments (eg. dev, staging and production), we will use different buildspec files and hence the need for using buildspec-prod and buildspec-otherenv files.</p><p>7. Now, the output jar file is uploaded to S3 and subsequent steps can download the same and use it as input for their builds</p><p>8. When the build runs, the build process role is <strong><em>code-build-vivcommonbuild-service-role</em></strong>. And this role needs to have access to S3 since this needs to be able to upload (and in some cases, download) artifacts from S3. Hence, we will need to provide permissions.</p><p>9. Hop over to the IAM service. Look for the role we are using for the build. In this case, look for <strong><em>code-build-vivcommonbuild-service-role </em></strong>in IAM.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B4QRjarpM_7Ru2LevGOjvQ.png" /></figure><p>10. And attach the <strong><em>AmazonS3FullAccess</em></strong> policy to this role as shown below. Note that you can make your access more restrictive by creating a custom policy that provides access to only the particular code build bucket and artifacts within</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qTIH0lbM0G6CY9e7t5CARg.png" /></figure><p>8. Click on Add Action to finish creating the action for building the first artifact. In our case we are done with the build for <strong><em>vivcommon </em></strong>library.</p><p>9. Next, we will create the build step for our second library, <strong><em>vivawsutils</em></strong>. vivawsutils essentially depends on two artifacts, one its source <strong><em>vivawsutilssrc </em></strong>and second, the <strong><em>vivcommonout</em></strong> artifact. And this is exactly the limitation that AWS CodePipeline/Build has.</p><p>10. The buildspec for <strong><em>vivawsutils</em></strong> looks as shown in the diagram below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/968/1*xA5jPa43qh2HdTzBzSv-_Q.png" /></figure><p>11. As part of the pre-build step, we download the dependeny jar from S3 as maven would have downloaded it and cached. The build code for <strong><em>vivawsutils</em></strong> picks up this jar (proxy artifact) from the maven repository cache and its source as input artifact and builds.</p><p>12. The set up of the build for <strong><em>vivawsutils</em></strong> works the same way as before and has to be set up exactly as we did for <strong><em>vivcommon</em></strong> except for the use of <strong><em>vivcommonsrc </em></strong>as input artifact.</p><p>13. Partial view of how <strong><em>vivawsutils </em></strong>build set up looks like and the resultant pipeline is shown in the next couple of imags.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*osPjtItB6EE1pyMfLzGxDQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/372/1*yJTLRg_YRyvcxRDSatM0Fg.png" /></figure><h3><strong>Conclusion</strong></h3><p>Our entire pipeline post the set up is shown below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*HDD1yLnsRIaW_T0HBW8hfg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9f2d22eb822a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/vivritiengineering/multiple-artifacts-in-aws-codepipeline-and-codebuild-9f2d22eb822a">Multiple Artifacts in AWS Codepipeline and CodeBuild</a> was originally published in <a href="https://medium.com/vivritiengineering">VivritiEngineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[enter, evaluate and execute]]></title>
            <link>https://medium.com/vivriticapital/enter-evaluate-and-execute-c01703612c95?source=rss-2c46741ffea6------2</link>
            <guid isPermaLink="false">https://medium.com/p/c01703612c95</guid>
            <category><![CDATA[fintech]]></category>
            <category><![CDATA[banking]]></category>
            <category><![CDATA[debt]]></category>
            <category><![CDATA[lending]]></category>
            <category><![CDATA[financial-services]]></category>
            <dc:creator><![CDATA[Yubi]]></dc:creator>
            <pubDate>Sat, 21 Jul 2018 09:55:31 GMT</pubDate>
            <atom:updated>2018-07-21T09:55:31.472Z</atom:updated>
            <content:encoded><![CDATA[<h4>The Vivriti Marketplace</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oSTETZIxmB5nYkFRHLWsug.jpeg" /></figure><p>The debt financing market in India is complex, characterised by a large number of counterparties — issuers and lenders / investors, vanilla and structured products, idiosyncratic needs and varying regulations. As a result, the market has been characterized by information asymmetry, relationships, manual execution, and inefficiency</p><p>This market stands at the cusp of being disrupted thanks to the “<strong>Vivriti Marketplace</strong>”, a marketplace for credit on tech for institutional borrowers to issue debt, discover investors, execute deals and raise capital efficiently and competitively.</p><p>In our journey of developing the Vivriti Marketplace we have taken a three-pronged approach: <strong><em>Enter, Evaluate, Execute.</em></strong> While the Marketplace will cater to all debt products — loans, bonds, commercial paper, asset backed securitisation, preference shares etc. — we use securitisation to illustrate our approach.</p><h4>enter — Automation and Self-Service</h4><p>Traditionally, securitization deals are executed through a heavily manual process that involves extensive communication between counterparties and stakeholders, sanitizing and processing the original pool of loans in arbitrary formats, aggregating the loans in the pool, removing risky loans, generating commercials/waterfall, and coordinating with our stakeholders till the final settlement of the deal. In a highly efficient manual process, this may take 4–5 working days.</p><p>In the Vivriti Marketplace, a client by initiates a securitization transaction by uploading two Excel sheets containing the loan pool details. Our clients use a varied set of information systems each with their own idiosyncrasies. Missing data, esoteric formats, extraneous data, badly formatted data, incorrect data are all par for the course. The onus is on us to parse, understand, sanitize, and normalize the data in the files. We use a combination of user input, online machine learning, simple and advanced algorithms, stemming, regular expressions matching, Levenshtein distance near matching, and bucketing of formats to process the files instantly and provide feedback to the clients on the size of the accepted pool, accepted and rejected loans, and reasons for their rejection. We have achieved 100% accuracy in processing files from existing clients and &gt;75% accuracy in processing pool files from new clients.</p><p>Pool selection is the act of cherry picking the pool to transform it, reduce risk and make it attractive for investors. In the Marketplace, we have used a simple greedy approach to perform pool selection with the click of a button within seconds. Further, our pool selection module applies regulatory filters and protects our clients from becoming non-compliant.</p><h4>evaluate — Analytics and Data Science</h4><p>Today, the structuring process is manual and is reliant upon an expert using his experience to structure the ideal Pass Through Certificates (PTCs). We intend to leverage public available and Vivriti proprietary data, mathematical models, artificial intelligence, and heuristics to develop intelligence into an automated selection and structuring process.</p><p>For our investors and clients to make informed decisions, we will provide data and tools including detailed client information, performance of the client’s past debt issuances, current market trends amongst others through the Marketplace. Investors can evaluate the credit worthiness of borrowers, their leverage, balance sheets, promoter, board and director information and Vivriti’s own evaluation of the client. Once investors make bids, the client can evaluate bids using tools and data available on the Marketplace.</p><h4>execute — Workflows and Coordination</h4><p>Private debt issuances are, by their very nature, customized transactions that are fine-tuned and executed after much parleying — over 160 steps with interdependencies. Several counterparties — rating agencies, legal firms, auditors, trustees, clients, investors and several internal teams, are involved and perform activities before deal consummation. Tracking the flow of a transaction is painstaking, error prone and laborious.</p><p>In the Marketplace, we have developed a configurable workflow engine to solve this. We call this workflow engine, the Tasks Infrastructure. We also built a componentized UI engine, which dynamically builds and displays the tasks, and their statuses to each of the counterparties.</p><p>The Marketplace has simplified the work of Vivriti teams, our counterparties, stakeholders, clients and investors and has helped them coordinate and execute better. We will also empower our clients and investors — with the help of data and machine learning — by allowing self-service pool selection to meet their specific liquidity and pricing needs / investment needs.</p><p>The Marketplace has simplified our securitization execution workflow reducing manual efforts, turnaround time and errors by 95%. Innovation in automation is a key for Vivriti to scale exponentially.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c01703612c95" width="1" height="1" alt=""><hr><p><a href="https://medium.com/vivriticapital/enter-evaluate-and-execute-c01703612c95">enter, evaluate and execute</a> was originally published in <a href="https://medium.com/vivriticapital">Vivriti Capital</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Technology @ Vivriti]]></title>
            <link>https://medium.com/vivriticapital/technology-vivriti-1d30c3c7e12d?source=rss-2c46741ffea6------2</link>
            <guid isPermaLink="false">https://medium.com/p/1d30c3c7e12d</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[debt-financing]]></category>
            <category><![CDATA[fintech]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[financial-services]]></category>
            <dc:creator><![CDATA[Yubi]]></dc:creator>
            <pubDate>Sat, 21 Jul 2018 09:50:02 GMT</pubDate>
            <atom:updated>2018-07-21T09:50:02.702Z</atom:updated>
            <content:encoded><![CDATA[<h4>Destroy the inefficiencies, we must!</h4><h3>Our work @ Vivriti</h3><blockquote>To continuously better the lives of all our clients through customized financial solutions</blockquote><h4>Objective @ Vivriti</h4><p>We founded Vivriti with the objective of creating the largest and most valuable marketplace for financial services serving a client base of financial institutions, corporates, small enterprises, and individuals that lack efficient access to financial services. The founding team is committed to making Vivriti Capital a one-stop shop for all the debt requirements of its target clientele.</p><p>The word, Vivriti means progress or development. At the root of development is capital. Our institutional clients, amongst others, constitute micro-finance companies and education loan companies. Our clients make progress possible by lending to poor entrepreneurs and small businesses who do not have access to capital through traditional channels. Our clients provide loans to students who pursue higher education and their dreams. We, at Vivriti, in turn, help our clients raise debt capital when they need it. With our retail business, we will also be lending directly to retail customers to pursue their financial and life goals.</p><h4>Our work @ Vivriti</h4><p>Companies target a specific capital structured that is dictated by regulatory requirements and the cost of such capital. Capital structure consists of <a href="https://corporatefinanceinstitute.com/resources/knowledge/finance/capital-structure-overview/">equity and debt</a>. We, at Vivriti, help clients raise debt financing. We have a multi stage approach with respect to our clients. We seek clients to partner with based on our understanding of the client’s sector. We develop a deep understanding of a client based on the client’s business model, and once we decide to partner with a client, we develop a relationship, one in which we keep ourselves aware of the client’s needs and work with them regularly to ensure they have access to capital they need when they need it.</p><h4>History @ Vivriti and elsewhere</h4><p>Our industry has long been characterised by</p><ul><li><em>Manual Processing </em>— We generate structured products, legal documents, term sheets, waterfall and other documents manually</li><li><em>Excel Usage — </em>We perform commercials, rick, credit and other financial computations with Excel, manually</li><li><em>Information Asymmetry — </em>Borrowers lack access to investors and critical information such as market pricing</li><li><em>Relationships — </em>Who you know rather than what you are and do dictates ability to raise capital</li><li><em>Concentrated Expertise </em>— Only a few “experts” can perform the kind of transactions we consummate</li><li><em>Subjective Decision Making — </em>Decision making, while still dependent on data, is also dependent on the expert’s experience, knowledge and views</li></ul><p>These issues have resulted in a less than optimal experience for our clients and investors</p><ul><li><em>Lack of Standardisation </em>— Different file formats and document structures make it harder for people to process and digest information</li><li><em>High Turnaround Time </em>— Time consumed standardising inputs, clarifying inputs and requirements, and processing input files manually all lead to high TATs</li><li><em>Sub-optimal Pricing and Investment Avenues — </em>Lack of transparency and data unavailability lead clients to take suboptimal decisions with regards to pricing. Undeserving borrowers may get better prices while deserving ones may get worse prices or my go unfunded</li><li><em>Nonscalability — </em>Impossible to scale rapidly to meet demand</li><li><em>Underemployment and Employee Morale </em>— Smart employees spend more time performing mundane tasks best left to machines</li><li><em>Errors, Reputation Loss </em>— Employee fatigue leads to mistakes in computation or documentation risking firm reputation</li></ul><h4>Approach @ Vivriti</h4><blockquote>“Over the past decades computers have broadly automated tasks that programmers could describe with clear rules and algorithms. Modern machine learning techniques now allow us to do the same for tasks where describing the precise rules is much harder” — Jeff Bezos</blockquote><p>We founded Vivriti with the sole purpose of disrupting this industry by leveraging technology as an enabler in solving these and many future problems. Engineering and technology is at the heart of our business. We take a multi-pronged approach towards solving the problems that plague our industry.</p><ul><li><em>Automation </em>— Eliminate all manual processes — computation of financials and commercials, generation of documents and waterfalls, collection of data from public and private sources, constructing structured data from unstructured data using techniques such as OCR and natural language processing (NLP), illustrate and illuminate through meaningful graphs (UI), manage and track performance of our clients and investor portoflios etc.</li><li><em>Data and Analysis </em>— Collect and store unstructured data from public and private sources in a data lake, provide business/users data at their fingertips, allow them to slice and dice, analyse and decide</li><li><em>Modelling, Evaluation and Prediction — </em>Develop several mathematical and financial models, use our data (primary and secondary) to evaluate investments, clients, macro and micro economic trends, interest rates, compute real time risk for our portfolios, build automated credit assessment modules, build default probability and loss estimation models, collateral/security values etc.</li><li><em>Process Management — </em>Develop workflows to coordinate and collaborate with our stakeholders, clients and investors, develop notifications and reminder mechanisms, provide information for decision making at the click of a button, develop chatbots and communication mechanisms to instantaneously communicate, clarify and make decisions</li><li><em>Authentication and Fraud Detection — </em>Use voice and face recognition for fraud detection, integrate with sources such as UIDAI, NSDL and CDSL for developing authentication systems of individuals and documents</li><li><em>Mobile Technologies — </em>Build mobility solutions such mobile branches, on the go marketplace access and decision making, customer touch points, on the spot loan approvals etc.</li><li><em>Payments and Post Transaction Handling — </em>Integrate with payment systems to allow funds transfer post transaction execution</li><li><em>Compelling and Convenient Dashboards — </em>Build compelling dashboards, graphs and UI with <a href="http://react.js/">React.js</a> to allow our users to easily interact with our data, analyse, make decisions and communicate it to us, counter parties and stakeholders all in one place</li></ul><p>We are doing this by building a single common online portal called <strong><em>The Vivriti Marketplace</em></strong></p><p>We are perfectly positioned to disrupt and lead this industry with our <a href="https://vivriticapital.com/whoweare/#governance">promoters</a> and <a href="https://vivriticapital.com/whoweare/#people">founders</a> who have, over six decades of experience in this field and have differentiated themselves with the knowledge capital they have built over the years. We are pioneers in this industry and have several firsts to our name including marquee transactions such the<a href="https://www.moneycontrol.com/news/business/economy/education-loans-become-latest-asset-class-in-indian-securitisation-market-2529059.html"> first education loans securitisation transaction in India</a> and social transformation transactions such as the <a href="https://www.livemint.com/Companies/g3TXWxBuz0E3UWaBLJX0DJ/Aye-Finance-raises-Rs25-crore-from-Hinduja-Finance-and-Intel.html">Aye transaction</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*QSgRLCWqzRm4p0gX.jpg" /></figure><p><em>Congratulations to the team that closed the first deal of the year for 200cr</em></p><h4>Values @ Vivriti</h4><p>We believe and abide by our core values of</p><ul><li>Customer Centricity</li><li>Integrity</li><li>Transparency</li><li>Work Ethic</li><li>Team Centricity</li><li>Passion</li></ul><h4>People @ Vivriti</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*3sY5HpMRwxilWf6-.JPG" /></figure><p><em>Team building at the offsite</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*AofsD5mCG5QU2DgU.JPG" /></figure><p><em>Lending a helping hand</em></p><p>Our people</p><ul><li>are out to make a difference and change the world</li><li>are smart, energetic, and enthusiastic</li><li>have strongly held opinions but loosely held convictions</li><li>want to work with and learn from other passionate developers</li><li>are diligent and persevering</li><li>want to solve complex and challenging problems</li><li>want to raise the bar on others and themselves</li></ul><p>And our engineers are also</p><ul><li>are passionate about coding and software development</li><li>are intrigued and inspired by new technologies and methodologies</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*fTPe84u3_9AW9n1j.JPG" /></figure><p><em>Solving problems together</em></p><h4>Fun @ Vivriti</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*dLKJFRRzWt62Yzxj.JPG" /></figure><p><em>Going for the kill</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*oqNu2_f0PcRAvl6N.JPG" /></figure><p><em>Preparing for the finals</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*zOmRmV9mAWpqpnd_.jpg" /></figure><p><em>VPL — I champions — Taj Savoy, Ooty, April 6, 2018</em></p><h4>You @ Vivriti</h4><p>If…</p><ul><li>You have a strong history of delivering results, <strong>and</strong></li><li>You have strong analytical, technical and communication skills, <strong>and</strong></li><li>You are passionate about software development, <strong>and</strong></li><li>You are keen on making history, <strong>and</strong></li><li>You have a strong sense of ownership, teamwork, purpose, and achievement, <strong>and</strong></li><li>You are keen on delivering products that help our customers, and play a meaningful role in the economy of the country, <strong>and</strong></li><li>You are smart and outcome focused, <strong>and</strong></li><li>You are keen on working the latest technologies and building world class products from scratch, <strong>and</strong></li><li>You have expertise/interest in building web applications, <strong>and/or</strong></li><li>You have expertise/interest in building enterprise scale back end systems, <strong>and/or</strong></li><li>You have expertise/interest in building cool front end applications, <strong>and/or</strong></li><li>You have expertise/interest in building applications on the cloud, <strong>and/or</strong></li><li>You have expertise/interest in building mathematical models, <strong>and/or</strong></li><li>You have expertise/interest in data science and analytics, machine learning and artificial intelligence</li></ul><p>then we</p><ul><li>are very excited about what you bring to the able, <strong>and</strong></li><li>would like to talk to you and explore this further</li></ul><h4>Workplace @ Vivriti</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*BLa8oy5NaT9J1qFM.jpg" /></figure><p><strong><em>Guests Welcome</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*XiDZ-XQmg2jcreBH.jpg" /></figure><p><strong><em>Meet, Greet and Colloborate</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*4W1djGRThgUb6ppZ.JPG" /></figure><p><strong><em>Welcome</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*obTmmWtmVwJ9ng_h.jpg" /></figure><p><strong><em>Work yet Relax</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*mw8KEUXYKwngf3KO.JPG" /></figure><p><em>Our Values in a nutshell</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*f0ItVuDut28xr70x.JPG" /></figure><p><em>The Big Board Room</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*cw-KiX_Y2eKtqTnR.JPG" /></figure><p><em>Conferencing with a view</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*MCUsQOgpYOVwR4PL.JPG" /></figure><p><em>Workplace with a view</em></p><h4>Stack @ Vivriti</h4><p>We are lucky to work on the latest and greatest of technologies whether it be cloud computing, front end, back end, machine learning</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*DMua3V9yjwtUf759.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*0PpWMcKlAH8iOdHu.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*D1jjUJKLSdi8MC-3.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/154/0*TCxFI5oqCLuqNAIm.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*muDXrvyETQ9Cxe6c.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*RYUxSaFAUceGlyHG.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/225/0*Sp4iVbVBu6DMiNL8.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/170/0*aZIopP5ZFt4DdmfL.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*0RHDhXm32ezChrke.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1d30c3c7e12d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/vivriticapital/technology-vivriti-1d30c3c7e12d">Technology @ Vivriti</a> was originally published in <a href="https://medium.com/vivriticapital">Vivriti Capital</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Microfinance Adapting To Changing Times]]></title>
            <link>https://medium.com/vivriticapital/microfinance-adapting-to-changing-times-2d85d9647b84?source=rss-2c46741ffea6------2</link>
            <guid isPermaLink="false">https://medium.com/p/2d85d9647b84</guid>
            <category><![CDATA[debt]]></category>
            <category><![CDATA[lending]]></category>
            <category><![CDATA[microfinance]]></category>
            <category><![CDATA[financial-services]]></category>
            <category><![CDATA[fintech]]></category>
            <dc:creator><![CDATA[Yubi]]></dc:creator>
            <pubDate>Sat, 21 Jul 2018 09:48:22 GMT</pubDate>
            <atom:updated>2018-07-21T09:48:22.861Z</atom:updated>
            <content:encoded><![CDATA[<p>I n the financial year 2016, the Indian microfinance sector finally appeared to have shed the baggage of the 2010 crisis. The sector had faced a significant test of its existence in 2010 when the government of Andhra Pradesh introduced an ordinance that effectively banned collections. In the resulting mayhem, liquidity dried up for the sector, aggregate assets under management (AUM) shrank by nearly 50% and the largest microfinance institutions (MFIs) were forced to seek cover under corporate debt restructuring. Smaller MFIs, those outside of Andhra Pradesh, spent the next few years in crisis mode for want of debt.</p><p>In 2016, however, evidence mounted that the shadow was lifting. The Government and the RBI awarded banking licenses to prominent MFIs across the country — a universal banking license to Bandhan and small finance banking licenses to eight MFIs. During the year, equity inflow was significant — and from a much wider variety of investors, debt issuances to mutual funds and insurance companies picked up, indicating in both instances wider acceptability among mainstream investors.</p><h3>At the same time, the cumulative AUM of the sector crossed INR 60,000 crores while portfolio at risk held steady at an average of 0.5% of AUM.</h3><p>The lending infrastructure had improved as well. Standard KYC, in the form of Aadhaar, was more ubiquitously available. Multiple credit bureaus had come up, improving the quality of data and end-customer assessment. MFIs had adopted technology at the last mile to enable digitisation of data at source, monitoring collection, controlling frauds and reducing turnaround times. Some were experimenting with cashless disbursements and collection models as also with mobile applications that could connect end customers with financial services such as deposits and investment products.</p><p>Amazingly, less than 18 months later, the sector has witnessed a fresh crisis. Growth has stalled in this period for NBFC-MFIs. Portfolio at risk is at 12% on an average, the highest since the 2010 crisis. Lenders have become nervous again, some have pulled back fresh exposure entirely. As with any crisis, questions on the viability or sustainability of the model have appeared again.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BTrUXI7cledFVP3XBxgpew.jpeg" /></figure><h4>The delivery of microfinance through private or non-governmental sectors appeared in India over two decades ago. The objective of employing the group-lending model was to enable access to credit for borrowers with no income proof in a cost-efficient manner with low credit loss. The fundamentals of the model were simple:</h4><ul><li>Keep ticket sizes low so that repayment amounts are low</li><li>Use social collateral as a surrogate for credit underwriting</li><li>Lend in groups to keep costs in check</li><li>Maintain frequent touchpoints with the client and instill process-discipline in both employees and borrowers</li></ul><p>The last two decades have seen several iterations and changes to corporate structures (culminating in advent of NBFC-MFIs and SFBs), as well as debates around concentration, scalability, and governance. The model itself has seen some tweaks as well — for instance, post the 2012 regulations governing NBFC-MFIs, several MFIs transitioned from a weekly repayment model to a monthly repayment model, ostensibly from the perspective of managing costs better, but perhaps without regard to impact on employee discipline as also higher installment amounts in an increasingly competitive environment that emerged a few years later.</p><p>Nevertheless, what is undeniable is that the group-lending model of microfinance, as practised by NBFC-MFIs, has been a resounding success. This is evident from the outreach indicators — MFIs have reached out to more than 5 crore customers in almost all states of the country, covering an overwhelming majority of districts. It is safe to say that barring geographies with difficult terrain, such as the upper reaches of the hilly north-eastern states for instance, MFIs have reached everywhere else in the country. It has not been an easy model to replicate either. Banks for instance, not including the MFIs who transitioned to banks, have not had nearly as much success in using the same model for delivery of credit. And it therefore follows that both NBFC-MFIs as well as the group-lending model of delivery of microfinance are here to stay and thrive as they deserve to.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rYU9T-P517M3CTqNtBxTNQ.jpeg" /></figure><h4>If lending models are to be categorised on first principles, one would broadly divide them into operational and credit models.</h4><ul><li>A credit-lending model is one that stresses on credit underwriting to manage delinquency, with a distinct credit team and centralised decision making. Small business lending/affordable housing finance would fall into this category.</li><li>An operational lending model is one where there is minimal underwriting of the credit risk. The focus is on distributing a standardised product and having a collections-led approach to minimising delinquency. Joint liability group lending/two-wheeler finance would fall into this category.</li></ul><p>Based on extensive field coverage spanning several districts at the peak of crisis, we feel that the group-lending model merits a closer look against the very characteristics that make it an operational model.</p><p>Core strengths of the group model — observations from the field:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CoJvSB4lSwSZonSR_LkElA.jpeg" /></figure><p>We conclude from the above that commodification of the microfinance product and employee, both of which came about in a bid to manage operating costs and both of which have played a huge factor in the growth of the sector in the years gone by, has in fact now made MFIs more susceptible to shocks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SPoLuimBu-dTXbEoPigyqQ.jpeg" /></figure><h4>Into the Future…</h4><p>We believe that it is time microfinance moved towards becoming a credit-lending model. We feel that the sector needs to move on from focusing on access to finance and look towards making this access efficient. In other words, MFIs should expend energy in identifying ways and means to understand the customer better and introducing nuance to the products on offer as well as managing the risk of this unsecured exposure. In doing so, MFIs will fundamentally start creating differentiation in the mind of the customer, which is critical from the perspective of retaining the customer.</p><p>Diversification of this nature, in our opinion, will entail the following:</p><ol><li>Embracing technology: The infrastructure afforded to microfinance institutions today is unparalleled. Borrowers with no income proof today create a digital footprint through a combination of Aadhaar, massive penetration of mobile telephony and their linkage to payments banks. With their regular connect with the client, MFIs are uniquely positioned to facilitate and create a habit of online transactions that will make this footprint deeper and more prominent. The advent of GST will, in the coming months, create a digital lock and key of the entire value chain or economic ecosystem of informal businesses. MFIs therefore sit on a goldmine of this massive volume of data, coupled with the data that they have already captured over the last several years on income sources (number and variety), residence stability, demographic change, etc.</li><li>Moving from a product view to a customer view:</li><li>Product design: MFIs can therefore improve their understanding of client behaviour in terms of repayment culture, income, expenditure and savings patterns — capturing the vagaries involved in the same. This understanding should feed into product design that optimises disbursements at the time the client needs the money the most and affords repayment options when the client has the money to repay. In theory, an MFI can therefore offer a whole range of products including working capital, consumption finance, crop finance, etc. These products can realistically match the cash flows of customers and not be a vanilla-term product.</li><li>Client retention: Demonstrating to the client a deep understanding of their ecosystem is a sure way of creating differentiation in the mind of the borrower and thereby retaining the borrower. Improvement in qualification and training of staff only adds further to the client’s confidence in the lender — as also better equips the MFI to handle stress situations — including political intervention. This is already evident in the fact that individual loan portfolios of MFIs which offer the product have done much better than the group lending portfolios even in worst affected states such as UP, Karnataka, and Maharashtra.</li><li>Revenue maximisation — Taking a client view therefore enables the MFI to offer nuanced need-based products. This can extend similarly to non-credit products as well including non-financial products.</li><li>Delivery: Digitisation of data and its optimal usage therefore facilitates underwriting a customer like never before — without adding substantially to costs. MFIs can therefore focus on hiring and training quality personnel whose sole purpose remains to put a face to the lender and maintain an individual relationship. The field intelligence that this staff can bring will only aid the underwriting. Additionally, the MFI does not require any more to deal with the client in groups and individual lending can take prominence.</li><li>Risk management: Risk management in microfinance has primarily been in the shape of monitoring or auditing the process. Recent events have demonstrated that this too is inadequate. Data-driven risk management is the need of the hour here as well. For instance, MFIs should use the massive intelligence and knowledge they have stored offline to help build a model that enables effective remote monitoring of event risks, draws patterns and perhaps, in due course, predicts where the next event risk is likely to occur.</li></ol><p>S everal MFIs have, in the past, tried to diversify product offering to their customers using what is commonly known as the “graduation” model. There have been varying degrees of success in managing portfolio quality in such models but very limited success in scaling this product up. We believe that the former is on account of inadequate or poor quality credit underwriting. In turn, this has been on account of the insistence of using the same employees who manage the group-lending model for credit in a bid to save costs. We feel that the lack of scale, in turn, has been on account of insufficient understanding of the client’s ecosystem, again on account of faulty data collection. Equally true is the fact that a substantial number of MFI borrowers use the loan amounts towards consumption and therefore graduating them to an individual loan is not prudent.</p><p>The digital approach we’ve outlined enables the lender to take a life-cycle view of the customer and offer different types of products at various points in time in this life-cycle. Therefore, if an MFI understands the customer’s economic ecosystem well, it may take an educated call to fund consumption. But fundamentally, the call should remain that of the MFIs.</p><p>There are, of course, challenges, foremost among which are regulations themselves, which prohibit “non-qualifying” assets exceeding a nominal proportion of MFI balance sheets. To remedy this, the regulator will have to take cognisance sooner or later, of the fact that the current regulations have killed innovation in microfinance in every respect except cost rationalisation.</p><p>In conclusion, we reiterate that the group-lending model of microfinance has done a phenomenal job in disseminating credit in remote pockets of the country. This is a job that it will continue to have to do. However, MFIs will need to ride the digital wave to understand individual customer better and not lose a customer to the group identity, retain the customer, offer multiple products, and more importantly, insulate their balance sheets from the bogey of event risk.</p><p><em>Originally published at </em><a href="https://www.vivriticapital.com/microfinance-2-3/"><em>www.vivriticapital.com</em></a><em> on August 31, 2017.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2d85d9647b84" width="1" height="1" alt=""><hr><p><a href="https://medium.com/vivriticapital/microfinance-adapting-to-changing-times-2d85d9647b84">Microfinance Adapting To Changing Times</a> was originally published in <a href="https://medium.com/vivriticapital">Vivriti Capital</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>