Speaker "ANSHUMAN GUHA" Details Back



Building ML Deployment Platform


AI/ ML is used across industries and while a lot of emphasis has been made on data preparation and building models, the actual model deployment is not as widely discussed. There are common challenges that companies across the industries face while deploying these models. In this talk, we will discuss some of these challenges and capabilities needed to build a ML deployment platform that provides a seamless and impactful experiences for all its stakeholders
Who is this presentation for?
NOTE: I am co-presenting in Siddharth Kashiramka for this same topic


I am working as Principal Data Scientist at Capital One in Card Acquisitions team. I have contributed to the ML model build pipeline and currently lead the deployment of COF financial models to customer facing platforms. Further, I helped to build credit risk & response Models using Deep Learning, NLP and GBM techniques with challenges including external economic impacts, evolving product offerings, changing customer behavior, ensuring model fairness & interpretability. These credit risk models have annual traffic of 10 – 12 million applicants via real time credit applications. Previously, I have worked as Data Scientist at SparkCognition. My role primarily involved developing anomaly detection/ prediction DS products using deep learning methods in IOT domain. Besides that, I have been working on AI based research initiatives like Transfer Learning, Domain Adaptation and Continual Learning. Looking forward to hearing from you.