Sep 2021 – Nov 2022 · 1 yr 2 mos

• Part of the founding team, Head of technology.
• As part of the founding team, I built lot of the core systems that power Byte, the AI tutor which helps students solve math problems and gives them step by step help when necessary. The systems that power Byte include question understanding modules for mathematical word problems and geometry problems (NLP and Vision), computer algebraic systems to solve the generated representations, a chatbot to interact with students and generate context specific conversations and hints by understanding student inputs, and systems that capture student’s knowledge and present it to the teachers.
• As the organization grew, I headed the technology team which built and maintains the teaching assistant which supplements the AI tutor. The teaching assistant helps teachers create assignments, manage their classes and allows their students to access Byte’s help when they are stuck. I lead the teams that are responsible for frontend, backend and dev-ops (deployment, observability and performance of all of Bytelearn's production systems). In addition, I worked closely with product, design, marketing and sales teams for scoping and prioritization of various features.

Jul 2016 – Aug 2021 · 5 yrs 1 mo

• Part of the founding team. Instrumental in developing the algorithms (Computer Vision and NLP) that power the suite of lung health products at Qure.ai. The algorithms' performance has been validated to perform on par or better than radiologists and published in top medical journals (including Nature, The Lancet) by independent research groups.
• I was extensively involved in conducting and evaluating large amount of deep learning experiments, building and maintaining the internal infrastructure to run the experiments and to store large medical datasets, setting up data pipelines. I also played a key role in getting the products deployed on resource limited embedded devices as well as on cloud infrastructure at scale.
• Apart from the above, I have contributed significantly to
• Various publications in conferences, radiology journals (including Nature)
• Multiple patents in AI for radiology domain
• Commercial partnerships with Health authorities, TB programmes, OEMs, Hospitals and teleradiology providers
• Research partnerships with top hospitals and universities across the world
• Clinical studies for evaluation of the algorithms and regulatory approval for the products
• Product Strategy, Hiring and team building
• Multiple other cross functional roles as a part of the start up

Jul 2015 – Jun 2016 · 11 mos