Job Description

Candiates with BS/MS degree in Computer Science or a related discipline with a strong understanding of algorithms and data structures from Tier 1 or 2 colleges only.

Responsibilities:

Productionize ML/AI models around semantic understanding and personalization which power our Search and Recommendation applications.

Building fast and scalable data pipelines aggregating various data sources.

Participate in lean and agile development processes in a test and monitoring-driven environment, while collaborating with Data Scientists and Product Managers.

Crunch historical data, search queries, product catalog, and images to extract hidden relations and features.

Conduct research to explore cutting edge ML / MLOps stack and ability to execute a quick POC.

Lead the effort to integrate and scale the ML components to a production-level that can handle terabytes of data.

Qualifications:

BS/MS degree in Computer Science or a related discipline with a strong understanding of algorithms and data structures (Preferably from Tier 1 or 2 college)

3+ years experience building ML-driven fast and scalable ML/analytical algorithms in a corporate/startup environment using Python, Spark/MLlib and Hadoop.

Experience working with distributed systems in a cloud-based production environment (AWS).

Strong Data Engineering & ETL skills to build scalable data pipelines. Exposure to data streaming stack (e.g. Kafka)

Relevant experience in fine tuning and optimizing ML (especially Deep Learning)models to bring down serving latency.

Exposure to ML model productionzation stack (e.g. MLFlow, Docker)

Excellent exploratory data analysis skills with ability to slice and dice data at scale using SQL in Redshift/BigQuery.

Decent understanding of various machine learning and natural language processing technologies, such as classification, information retrieval, clustering,knowledge graph, semi-supervised learning and ranking.

Excellent problem solving, analytical and communication skills.

Exposure to deep learning stack (PyTorch/Keras/TensorFlow) and techniques

(Representation/Transfer Learning, RNN/LSTM, Transformers)