TechGig: Best Practices for AI/ML Model Development, Deployment and Usage | ft. Kasthuri Rangan Bhaskar

  • Share

3
  < 1 minute

In a webinar hosted by Techgig, Kasthuri Rangan Bhaskar, Vice-President – Products, BCT Digital, shared his insights on the extensive use of AI/ML models in the financial industry. They are used for customer sourcing, credit scoring, campaign management, fraud risk management, anti-money laundering, risk management among others. ML models have enabled and accelerated tasks that otherwise would have been cumbersome, manually time-consuming or even impossible. For example, in fraud risk management, ML models can identify patterns of suspicious transactions from huge volumes of transactional data, which would not have been evident through casual observation. The use of ML models can also bring about tremendous process and cost efficiencies.

However, there is a flipside to the usage of ML models. These models could suffer from serious flaws such as instability, bias, and explainability, if they are not constructed, used, and governed properly.

Key points of discussion

  • AI models and their usage in fintech
  • Understanding key risks across the ML model lifecycle
  • Functional and technical challenges in deployment at scale in an Indian context
  • Consequences of such risks
  • Key regulatory concerns and guidelines – Bias and explainability
  • Best practices to be followed across the model lifecycle – fundamental pillars
  • Illustrative use cases in real life

Speakers

Kasthuri Rangan Bhaskar

Vice President – Products, BCT Digital

Kasthuri Rangan Bhaskar
Vice President – Products, BCT Digital