Aug 2023 – Nov 2024 · 1 yr 3 mos · Greater London, England, United Kingdom

• Role: Own end-to-end engineering development working with ML researchers, setting software and data practices for the AI product.
• Designed, optimized, and built a large-scale distributed system for a preprocessing pipeline in an AI product. This involved engineering efficient solutions to parallelize data processing of up to 500 million compounds over dynamic cloud-based spot and on-demand instances with the creation of a datastore, reducing compute cost by 50%, making running time <1/8th of the original, increasing request throughput by 7x, and ensuring robust monitoring and productionization.
• Engineered a cost calculation ETL product to precisely allocate AWS EC2 compute costs to individual workflows and user request granularity, significantly enhancing cost transparency and efficiency across the platform.
• Built a library for a computational platform to efficiently map >10TB 3D molecules using NoSQL and Redis caching.

Aug 2022 – Aug 2023 · 1 yr · Greater London, England, United Kingdom

• Role: Technically lead a team of 4-5 people with a product manager to build products; Mentor new grad software engineers.
• Designed and shipped a ward management product for UK hospitals by integrating hospital EHR data into a unified platform. This helped reduce inpatient waiting list by 28% improving discharge efficiency, and saving lives in the process.
• Led end-to-end development of an AI-powered cohorting system to analyze medical data and recommend inpatients for virtual wards - freeing physical beds, reducing readmission risk, and enabling clinicians to manage larger patient cohorts safely. Part of NHS England's national virtual ward expansion to 10,000+ beds.
• Built London's remote patient monitoring hub software. This centralized data from multiple hardware providers,
• enabling real-time alerting, staff assignment, and risk assessment. Pilot converted to ~$1M statement of work.

Dec 2021 – Aug 2022 · 8 mos

• Role: Build backend services and scalable data pipeline for multiple health tech product
• Designed and developed payouts microservice by integrating RazorPay payment APIs using Flask framework, Python libraries like Sqlalchemy, and webhooks to automate the payment transactions for all online consultations and other products company-wide, resulting in secure and instantaneous payouts to the doctors.
• Transitioned the core database to Redshift by designing and implementing a scalable data pipeline using ETL tools like AWS Glue and WorkFlow with Spark. This helped in reducing average cost by 40% and saving daily cron time by 2.5 hours. This handles huge data volumes (~100 million rows) and is intended to be used for all company products.
• Built Offline EMR product with microservices-based architecture, utilizing sync and edge services to enable both offline and online mode simultaneously with the help of schedulers and no-SQL databases like MongoDB.
• Developed 4 products that involve data processing and data management solutions using complex data of more than 15 million distinct patients, to provide valuable insights leveraged by more than 6000 doctors and giant pharma firms.
• Created a referral service product using Golang and gRPC by integrating various third-party services to allow existing stakeholders to onboard new referrals, thereby providing credit reward benefits for both parties.

Nov 2020 – Dec 2021 · 1 yr 1 mo

Jul 2019 – Dec 2019 · 5 mos · Greater Delhi Area

Dec 2018 – Apr 2019 · 4 mos

• - Developed a machine learning pipeline for NLP in Athena AI (interview bot). This will be used for technical interviews in the company's hiring process.