Data Scientist II- Time Prediction

Key Skills

Apache Hadoop, C Programming, C++, Data science techniques, Java Programming, Machine learning techniques, MapReduce, Python Programming, PyTorch, R Programming, Scala Programming, SQL, TensorFlow

Job Description

Within Data Science, the Marketplace Fulfilment Platform (MFP) DS team builds data science products to optimize delivery efficiency and power post order delivery experience for both customers and delivery partners.

We build and own low-latency, high-throughput models powering time predictions, routing and dispatch optimization across different business lines of Swiggy. We also work on models for behavior of customers and delivery partners, spatio-temporal community detection, graph learning and Swiggy Simulation platform that span econometric, statistical, machine learning and deep-learning approaches.

We own or co-own several initiatives with a direct line of sight to business impact. We are a group of data scientists, operations research scientists and ML engineers with background and expertise in ML/DL, optimization, statistics, data mining, econometrics, graph theory and large-scale distributed systems.

Key Problem Statements

- Enhance Assignment to Delivery time prediction model to enable experiments with optimization used in batching and assignment - Enable experimentation with Service recovery delay

What You’ll Be Doing

• As a Data Scientist II in the MFP team, you will be building models that enable Swiggy to deliver on our promises with high efficiency and best experience. You will work closely with world-class Engineering/Product/Operations teams to build state-of-the-art solutions that deeply impact Swiggy’s business and customers. 
• You would be owning a significant portion of the charter to drive delivery efficiency and delivery experience via deployment of state of the art models for dispatch optimization
• You would be collaborating with product and engineering to design and document an ambitious strategy, identify key metrics to drive, support with proof of concept, develop and execute on prioritized ideas, experiment, scale to achieve direct impact to Swiggy’s business
• You would be working with and mentoring fellow team members on the owned charter.
• You would be contributing to the brand of the organization by publishing blogs, technical papers and presentations at conferences

What we look for

• Masters/PhD with focus in areas like Computer Science, Applied Mathematics, ML/DL, Statistics, Operations
• Strong depth and breadth in machine/deep learning (theory and practice), optimization methods, data mining, statistics and linear algebra
• Hands-on programming experience (intermediate to advanced level PySpark, Python, Scala, SQL, Keras/Tensorflow) on big data (familiarity with AWS and open-source ecosystem tools) with the ability to manipulate structured and unstructured data
• Strong spoken and written communication skills, able to explain complex ideas in a simple, intuitive manner, write/maintain good technical documentation on projects
• Experience in mentoring a small team of data scientists preferably in a startup or tech company
• Experience with shipping near-real time inference ML data products in an engineering organization interfacing with other teams and departments to deliver impact.