Alexa

50 Best Companies to Work For 2018

Implementing artificial intelligence in crime fighting: Tookitaki Holding Pte Ltd

thesiliconreview-abhishek-chatterjee-ceo-tookitaki-holding-pte-ltd-2018When it comes to the financial services sector, banks and other financial institutions often have a difficult time trying to monitor and prevent any fraudulent activities. The larger an institution becomes, the more the customers it handles and as a result, chances of illegal trading increase proportionately. Governments pass legislation after legislation to curb such practices but have little success since illicit banking practices are hard to detect and even harder to prevent. A bold new venture is attempting to employ machine learning algorithms to help banks and financial institutions ensure full compliance with government regulation and combat financial crimes.

Incorporated in November 2014, Tookitaki started off by building a machine learning platform that offers higher accuracy and faster time to market for predictive analytics. The founders of the newly conceived firm soon realized the size of the market for financial regulatory compliance. Thus, Tookitaki set out to build top of the line applications that cater to anti-money laundering and reconciliation efforts.

Tookitaki provides enterprise software solutions that enable financial institutions to create sustainable compliance programmes. As a mission, it aims to improve risk and efficiency through machine learning enabled software solutions, which are actionable, auditable and scalable. In the regulatory compliance space, the company’s offerings cater to anti-money laundering and reconciliation spaces.

Tookitaki’s range of products enables banking institutions to ensure sustainable compliance

Introduction to Anti-Money Laundering Suite

Today, digitization is a global phenomenon and the amount of ‘wire’ activity is growing significantly, making it difficult for financial institutions to monitor and detect suspicious transactions and names effectively and efficiently. Additionally, criminal activity is increasingly sophisticated and terrorist financing uses cutting-edge technology to circumvent detection. As a result, regulators enforce a higher level of scrutiny in the money laundering space. Therefore, financial institutions require advanced technological solutions to address its AML risks.

Tookitaki’s anti-money laundering suite is a one-stop solution to transaction monitoring, names screening, and customer risk ratings. The firm focused on modules for transaction monitoring and names screening to help prioritize alerts and subsequently bring in a significant reduction in false alerts. The Anti-Money Laundering Suite complements existing systems and applies machine learning techniques to learn patterns from historic cases, recent transactions and sanctions list updates to create a sustainable alerts management framework. The AML suite improves operational efficiency by 40%.

Introduction to Reconciliation Suite

Most of the recon solutions in the market are RPA providers or legacy rules-based solutions. RPA or rules-based applications fail to handle exceptions as they are based on a set of conditions, while exception handling requires human judgment beyond the basic set of rules.

Tookitaki’s Reconciliation Suite offers machine learning-powered end-to-end enterprise reconciliation software. It has modules on matching, exception handling and adjustment amount recommendations. It learns from historical patterns and can recommend, matching, exception resolution and adjustment amount, without any human intervention. The suite offers more than 90% accuracy in matching and exceptions predictions.

Using machine learning to accomplish bold objectives

Tookitaki positions itself as an enterprise play bringing machine learning in regulatory compliance for the banking & financial services industry.

The business objective in AML (Anti-money Laundering Suite) is primarily twofold – to reduce false alerts and to improve risk (detection of missed STR or Suspicious Transaction Report). Tookitaki does both using machine learning.

The business objectives in Recon are – bringing automation in exceptions management and reducing cost. Tookitaki undertakes both using machine learning.

Some of the company’s largest clients are Standard Chartered PLC and Société Générale S.A.. It mainly focuses on the banking and financial services sectors to curb illegal trading and help authorities detect malpractice.

Factors that give Tookitaki the coveted edge over its competition

All the solutions that the company provides have five distinct traits that make it a market leader:

  • Actionable: Outputs for AML and recon are combined with relevant insights for business users to take quick action.
  • Interpretable: Machine learning models were considered to be a BlackBox. Tookitaki changed the paradigm by providing detailed algorithmic transparency by explaining models through dependency path and important features. This is combined with powerful analytics for each domain.
  • Scalable: The company’s engine or data science studio (TDSS) that powers both applications of AML and Recon can seamlessly connect to any bank’s upstream and downstream systems for easy production deployment. Tookitaki offers both batch and real-time processing of outputs.
  • Commitment: The company guarantees 30% reduction in false alerts for AML and 90% plus accuracy for matching and exceptions for recon.

Client testimonials speak volumes about Tookitaki’s products

Reconciliation Suite

A premier European multi-national banking and financial services company deployed Tookitaki’s Reconciliation Suite for GL reconciliation. The company offered the system complementary to the incumbent matching engine. The system used machine learning to improve exception handling. All the exceptions were predicted with 99% accuracy. Additionally, the Reconciliation Suite also generated rules to help investigators understand exceptions with such a level of detail and clarity that it reduced investigation time by 40%. The automatic learning reduced time and subjectivity, bringing significant operational efficiency in GL reconciliation.

Anti-Money Laundering Suite

A leading regional bank in the United States wanted to improve the efficiency of the current transaction monitoring system. The current system generated over 95% false alerts. It was extremely difficult for the existing team of analysts to dispose alerts on time, resulting in significant backlogs. The bank nearly partnered with a BPO company to solve this. Tookitaki presented its AML suite and convinced the bank to try a pilot on the transaction monitoring module. The objective was to reduce false alerts by 30% and explain prediction results to the regulator during the periodic review process.

Tookitaki carried out the pilot on the commercial banking segment and demonstrated a 40% reduction false alerts on test data and was also able to give a demo of the solution and explain the prediction results successfully to the regulator.

Meet the man behind Tookitaki’s success

Abhishek Chatterjee, CEO

Mr. Chatterjee has a master’s degree in Applied Mathematics from the University of Southern California. Before founding Tookitaki, he was a quantitative trader in JP Morgan, New York. He also worked in DoubleClick, a subsidiary of Google as a software engineer. He loves to solve puzzles and is an avid reader. On a lighter note, he has a killer follow-up strategy with his clients and peers.

“Commitment to clients is our obsession. We want to build a platform that helps clients take distinctive decisions that leave lasting and substantial improvements in their performance.”