Suju Rajan

Deep Personalization in Jobs Marketplace: A LinkedIn Perspective.

At LinkedIn, our jobs marketplace attempts to optimize the matching of hiring managers with the job seekers with the goal of getting more job seekers hired. In this talk, we will look at how we use deep models to capture job seeker’s career trajectories and hiring managers requirements, use of multi-task learning to optimize for both sides of the marketplace while driving equitable outcomes across the board.

Suju Rajan is a Senior Director at LinkedIn, where she leads the Standardization and Enterprise AI team at LinkedIn, reporting directly to Chief Data Officer Igor Perisic. Her team joins together all of the work done in Enterprise and Standardization AI at LinkedIn. Her team's work powers the hiring, marketing, and insights products that the company's customers rely on every single day. The speed and accuracy with which LinkedIn provides enterprise solutions (for learning, hiring, sales, etc.) relies heavily on well-constructed taxonomies of all of its Economic Graph data. By combining the Standardization team into Enterprise AI, her team tightens the feedback loop between the company's members/customers and their solutions, to improve our data and recommendations.

Suju is also passionate about research. She and her team are exploring many innovative projects at LinkedIn in areas like applied deep learning, interesting problems in marketplace optimization, and large scale AI productivity.

Finally, she is personally passionate about the ability of AI to make society more equitable and improve people's lives. As a woman in the male-dominated tech industry (and the even more male-dominated field of AI), she is an advocate for women in the field of AI and is interested in showing ways that AI can be used to mitigate social biases, rather than amplifying them.

Previously, she worked as the head of AI at Criteo and began her career as a researcher at Yahoo! Labs.

Buttonlinkedin
This website uses cookies to ensure you get the best experience. Learn more