Apr 2022 – Oct 2022 · 6 mos · Gurugram, Haryana, India

• Led the evolution of DotPe’s core commerce platform, owning system design and delivery across ordering, pricing, growth, and analytics systems serving large-scale SMB storefronts and high-throughput transaction flows.
• ● Owned the architecture and scalability of backend services powering the end-to-end ordering lifecycle, including cart, checkout, offers, delivery logic, and store management. Designed backend-driven APIs to support Android migration to Kotlin, enabling consistent platform behaviour across clients and improving long-term maintainability.
• ● Architected a real-time cart, pricing, and promotions engine and led the launch of a premium subscription platform, driving a 19.7% increase in customer acquisition and an 11.3% uplift in paid conversions. Built flexible coupon and promo systems to support rapid experimentation by product and growth teams.
• ● Established platform-wide API security standards using AES-256 encryption, rate limiting, and middleware-based access controls. Improved system latency and throughput through caching strategies, database indexing, pagination, and concurrent request handling, enabling reliable performance under peak traffic.
• ● Led third-party integrations across KYC, Shiprocket logistics, and Truecaller, designing resilient, observable integration layers. Built a self-hosted web analytics platform using Matomo, enabling store-level behavioural insights and operational visibility at scale.
• ● Mentored engineers, led design reviews, and partnered closely with product and customer success teams to translate business requirements into scalable, production-grade system designs and long-term platform roadmap initiatives.

Jan 2021 – Mar 2022 · 1 yr 2 mos · Gurugram, Haryana, India

• Contributed to building the foundational backend services that powered DotPe’s merchant and ordering platform, focusing on correctness, scalability, and developer-friendly API design.
• ● Implemented the cart, checkout, and delivery calculation engines from first principles, including validation frameworks, tax and fee computation, and order consistency guarantees across distributed services.
• ● Developed key platform capabilities, including merchant onboarding flows, role-based multi-user access control, store configuration tools, and operational notifications, enabling rapid SMB adoption and internal support efficiency.
• ● Built an admin console for customer success and operations teams, streamlining merchant issue resolution, store lifecycle management, and platform configuration without direct engineering involvement.
• ● Improved backend performance through query optimisation, structured caching, indexing strategies, and pagination standards, ensuring stable API response times under growing traffic and merchant load.

Jun 2019 – Dec 2020 · 1 yr 6 mos · Noida, Uttar Pradesh, India

• Led the design and delivery of multiple intelligent systems across internal engineering platforms and consumer-facing device features, operating at the intersection of machine learning, system architecture, and cross-team technical leadership.
• 1. PLM Score (Engineering Productivity Platform)
• ● Owned the end-to-end design of an ML-driven triage and quality scoring system for Samsung’s internal bug management platform, addressing long-standing inefficiencies in issue ownership and duplicate defect resolution across Android device lines.
• ● Built a classification and clustering pipeline over historical PLM data to automatically route issues to the correct owning teams, identify defects solvable through a single shared change, and assess developer effort and code quality signals. Improved resolution accuracy to 76.4%, significantly reducing assignment latency and duplicate engineering work across multiple product teams.
• 2. On-Device Face Recognition (Android Go Platform)
• ● Led the development of a lightweight face recognition system optimised for low-spec, low-sensor Android Go devices, balancing model performance with strict compute, memory, and camera quality constraints.
• ● Designed and fine-tuned the model pipeline to operate reliably under real-world conditions, achieving 85.3% precision while maintaining on-device performance and power efficiency targets.
• 3. Behavioural Biometrics (User Security & Privacy)
• ● Architected a swipe-pattern behavioural authentication system to enhance device security beyond static unlock patterns.
• ● Designed feature extraction and model training strategies to capture user-specific behavioural signals, enabling continuous authentication and mitigating shoulder-surfing risks. Delivered a system achieving 71.2% precision, with a strong focus on user privacy, on-device processing, and scalable deployment across device variants.

Jan 2019 – May 2019 · 4 mos · Noida, Uttar Pradesh, India

• Led the design and implementation of an Android malware detection pipeline focused on identifying malicious behaviour through large-scale permission analysis, addressing security risks introduced by modified and unofficial APK distributions.
• Built and trained a classification system over a corpus of 140,000+ applications, mapping each app’s behaviour against a standardised framework of 330 predefined user permissions. Designed the feature engineering and model training strategy to learn expected permission patterns across application categories and flag anomalous behaviour in previously unseen apps, achieving 79.6% classification accuracy.