Sr. Data Scientist

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

Upstox is one of India's leading Fin-Tech companies with a mission to simplify trading & investing to make it easily accessible to the masses. We aim to enable everyone, from new investors to seasoned traders, to invest across multiple categories with our state-of-the-art trade & investment platform and commission-free pricing. We offer numerous asset categories to invest in, like Stocks, IPOs, Mutual Funds, and more.

By focusing on our customers’ needs and equipping them with personalized yet powerful tools, we witnessed a steep growth of 800% in our customer base from 25 Thousand in 2017 to 2 Lakh in 2019. With 1500% growth in 2020, currently, over 3 million customers trust us with their investment decisions, thus setting us on the course to become an industry-leader in the country. 

Our mission is simple - to break down the complexities of investing and make it more effortless, accessible, affordable, and easy for the masses to adopt. This key principle when infused with intuitive design and leading-edge technology will help us empower every Indian to take control of their investments.

RKSV Securities was founded by Ravi Kumar and Shrinivas Viswanath in 2009 and was soon graced by Kavitha Subramanian as the third co-founder in 2016. Backed by Ratan Tata, Upstox had raised $4 million in Series A funding in early 2016 that was led by Kalaari Capital. The Series B funding round scaled multifold and witnessed an investment of $25 million by US-based investment firm - Tiger Global Management in September 2019.

We have a team of highly skilled technology and finance professionals, and are currently looking for highly motivated field experts to be part of our high-energy team.

Role - Lead Data Scientist / Data Science - Lead (7+ years core data science experience/5+ for exceptional Candidates)

Responsibilities:

Upstox is looking to hire exceptional data scientists. The ideal candidate would have the ability to build sophisticated ML systems within complex environments that help drive key customer & business objectives through automated data-driven decision-making.

1. Understand the business and data environments in-depth, to identify opportunities to leverage ML and drive business and customer success. Work with the Product, Business & Design teams as well as the BI & Analytics team to cearly draw up business charters with a priori impact analyses.

1. Architect ML systems within which ML solutions are to be embedded and design the interaction with Upstox’s large data assets. The ability to work closely with ML & Data Engineering teams and speak the language at these interfaces. 

1. Evaluate and select, based on previous experience and a close understanding of the SOTA in relevant ML sub-fields, algorithms, tools and technologies appropriate for delivering the core ML solution for the problem at hand. 

1. Train & test algorithms for accuracy, generalizability, scalability, fairness, explainability and stability. Design, implement and maintain domain-specific feature stores & monitoring thereof.

5.Execute and drive execution in a cross-functional, multi-system setting with a clear view of configurability and flexibility in integration. Ability to adapt ML systems to availability of real-time or offline data sources.

6.Design & implement monitoring systems that measure key aspects of performance, including model or data drift, and out-of-time accuracy deterioration. Implement hybrid learning systems & annotation systems.

7.Design & implement model lifecycle management systems including AutoML-based retraining pipelines, champion-challenger model systems, and automated model redeployment basis key performance differentials between currently deployed models & retrained models.

Required Skills and Experience:

• Deep understanding of data, algorithms and engineering are key to this role. Ability to write exceptional, high performance code (in any data science language; Python preferred) is essential. Demonstrable track record of using DS & ML for achieving business goals and objectives is highly desirable. 
• Specifically, the candidate should have exposure to the following ML & DS sub-fields:
• Recommender systems and news feed personalization
• NLP, sentiment analysis, text classification, NER. Should have implemented models in production in at least one DL framework (PyTorch, HuggingFace or TF/Keras). 
• Performance marketing optimization and cross-channel attribution modelling. 
• Pricing optimization & dynamic, behavioral experimentation
• Customer segmentation & CLTV modelling
• MLOps and low latency inference 
• Forecasting at scale

It is not expected for the candidate to have exposure to all of these domains, and wherever possible, depth is preferred over breadth. The ideal candidate would have implemented models at the SOTA of the sub-field to which they have experience.

The ideal candidate will be a cloud-native (preferably AWS) and have team & project management experience.