{"id":939,"date":"2022-10-19T16:23:18","date_gmt":"2022-10-19T16:23:18","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/startups\/blog\/?p=939"},"modified":"2024-11-04T14:17:15","modified_gmt":"2024-11-04T22:17:15","slug":"qard-idea-to-mvp","status":"publish","type":"post","link":"https:\/\/www.microsoft.com\/en-us\/startups\/blog\/qard-idea-to-mvp\/","title":{"rendered":"How Qard went from idea to MVP – #LaunchWithAI"},"content":{"rendered":"\n
For this post in the #LaunchWithAI series<\/a>, I met with Azzeddine Chaibrassou<\/a>, founder of Qard<\/a>, a fintech startup, based in Paris, France, innovating in credit risk management. Qard started by addressing a niche pain point felt by many in the industry and took it to the next level moving swiftly from idea to MVP, all while launching the startup with their core in AI. Read on to find out what we discovered.<\/em><\/strong><\/p>\n\n\n\n “The main idea behind Qard was to give access to reliable information about companies to fintech, brokers, venture loan companies, neobanks, and other financial institutions. Today, disclosure of financial information about companies is registered as a legal act in France and other countries around the world. Yet, not many products existed that would facilitate bringing this information meaningfully to the consumers, to make the data actionable. Qard solves this problem by facilitating connections to various data sources, collecting relevant information, and bringing them together, either through an API or a dashboard.”<\/p>\n\n\n\n “Going from idea to execution and MVP is one of the most resource and time intensive phases for a startup. Technologically, our product goal was to source and scan millions of documents, detect patterns and signals about the financial health of companies from the data set and eventually relay these patterns to our customers with our APIs. From the get-go, we\u2019d have needed recruitments efforts to get the right DevOps and data engineering pipeline, and the in-house development would\u2019ve taken months.<\/p>\n\n\n\n Having led teams through such 0 to 1 product creation, our head of engineering, with more than 15 years of industry experience, directed us towards leveraging high-performance Azure models to accelerate our time to development. So, we adopted and used Azure’s tools, which saved us a lot of time. Having tools dedicated to what you want to achieve saves time in technical development, but also in product management. Using ready-to-use solution Azure APIs vs installing and maintaining in-house ML pipelines freed the time for our data scientists to manage more complicated projects.”<\/p>\n\n\n\n “Qard uses natural language processing (NLP) to extract and explain useful information from the millions of documents available. From this data, we detect patterns and signals about the financial health of companies and relay this information to our customers.<\/p>\n\n\n This process, overall, has the following major components:<\/p>\n\n\n\n Through this process, <\/span>we\u2019ve<\/span> already served tens of thousands<\/span> of PDFs<\/span>.”<\/span><\/p>\n\n\n\n “Our API now provides<\/span> information and data processing like never before.<\/span> Our<\/span> technical team <\/span>still consistently faces<\/span> optimiz<\/span>ation<\/span> challenges. <\/span>Our Microsoft mentor has been helping us work through these challenges with which we continuously improve<\/span> and A\/B test to <\/span>establish<\/span> our product-market fit.”<\/span><\/p>\n\n\n\n Visit <\/span><\/span>Launch <\/span>With<\/span> AI<\/span><\/span><\/a> for one-pager on how to get started<\/span> on your building journey with <\/span><\/span>Microsoft for Startups Founders Hub<\/span><\/span><\/a>.<\/span><\/span> <\/span><\/strong><\/em><\/p>\n","protected":false},"excerpt":{"rendered":" Using ready-to-use Azure \u0003APIs vs installing and \u0003maintaining in-house ML \u0003pipelines freed our data \u0003scientists to manage more \u0003complicated projects.<\/p>\n","protected":false},"author":20,"featured_media":943,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ms_queue_id":[],"footnotes":""},"categories":[57],"tags":[19,253,130,234,233],"coauthors":[],"class_list":["post-939","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-prototyping","tag-azure","tag-databricks","tag-launchwithai","tag-named-entity-recognition","tag-ocr"],"yoast_head":"\nHow Qard got started<\/h2>\n\n\n\n
<\/figure>\n\n\n\n
Moving from idea to MVP<\/h2>\n\n\n\n
Leveraging NLP for the Qard solution<\/h2>\n\n\n\n
<\/figure><\/div>\n\n\n
\n
Taking your MVP to market<\/h2>\n\n\n\n
Accelerate your journey<\/h2>\n\n\n\n