The data science team is responsible for solving business problems on complex data. Data complexity could be characterized in terms of volume, dimensionality and multiple touchpoints/sources. We understand the data, ask fundamental-first-principle questions and apply our analytical and machine learning skills to solve the problem in the best way possible. The problems that we tackle are focused on our 2 products/IP related to anti-money laundering (AML) and financial exception management. In addition, the team plays a pivotal role in shaping company IP, primarily in area of approach, model selection, model tuning and feature
engineering, so that engineering can take this up and automate the same via a pipeline
Our Ideal Candidate
• Identifying the key problems faced by our clients and the AML product, and executing a research plan to solve it;
• Implementing solutions to upgrade feature engineering and modelling pipelines of the AML product;
• Maintaining effective communication with the product, delivery and engineering teams, and providing sufficient assistance based on priorities
Requirements:
• Prior experience in Python and Spark;
• Ability to execute research plan using machine learning technologies;
• (Excellent problem-solving skills;
• Expertise in NLP, graph database and graph algorithms would be a plus.
• Deep experience in big data would be a plus.
• Relevant work experience of 3-5 years
AMLS focus areas for the role
• Name Screening
a. Building NLP pipeline to extract essential profile information from free text;
b. Finding the best solution to do entity resolution;
• Transaction Monitoring
c. Upgrading the unsupervised pipeline to do more effective clustering and anomaly detection;
d. Building a link analysis pipeline using graph databases and algorithms;
e. Upgrading the supervised pipeline by improving on feature engineering and making clustering more effective
Desired Non-technical Requirements
• Very strong communication skills both written and verbal
• Strong desire to work with start-ups
• Must be a team player
IF YOU LIKE TO HAVE AN OPPORTUNITY HERE, DO CLICK "WANT TO VISIT" AND APPLY. ONLY SHORTLISTED CANDIDATE WILL BE CONTACTED.