Nov 2021 – Mar 2023 · 1 yr 4 mos

Jan 2021 – Nov 2021 · 10 mos · India

• Architected and developed a self-sustainable Enterprise Resource Planning (ERP) system for a global e-commerce infrastructure, supporting multi-market operations.
• Engineered a "Zero-Ops" ecosystem, automating infrastructure management and deployment pipelines to eliminate the need for a dedicated DevOps team.
• Streamlined international expansion by building scalable modules that allowed the rapid onboarding of new markets from scratch.
• Delivered a user-centric administration platform, enabling non-technical stakeholders to manage complex system operations without engineering intervention.
• Optimized legacy workflows, developing custom solutions to automate day-to-day business logic and reduce manual overhead.

Oct 2020 – Jan 2021 · 3 mos · India

• Engineered a high-throughput data ingestion and processing pipeline using Scala and Apache Spark to analyze user behavior on an e-commerce platform.
• Implemented an event-driven architecture using Apache Kafka, enabling real-time capture and streaming of user interaction logs (clickstream data).
• Developed complex data transformation jobs to convert raw click data into actionable insights regarding user retention and product engagement.
• Leveraged the Spark Structured Streaming API to handle massive datasets with low latency, ensuring scalable performance during peak traffic.

Sep 2020 – Dec 2020 · 3 mos · San Francisco, California, United States

• Engineered an end-to-end synthetic data generation pipeline using Blender and Python, automating scene rendering, ground-truth annotation, and occlusion calculation to overcome dataset scarcity.
• Developed a custom Python-to-Blender API wrapper, abstracting complex 3D rendering tasks into a programmable framework for automated dataset creation.
• Implemented Sim-to-Real domain adaptation techniques to translate synthetic 3D renders into photorealistic images, bridging the "reality gap" for computer vision models.
• Optimized a CycleGAN architecture by modifying loss functions and training protocols, achieving higher feature fidelity and color realism compared to baseline models.
• Conducted extensive ablation studies on model architectures and training sets to maximize the visual quality of style transfer.

May 2019 – Aug 2019 · 3 mos

• Architected a privacy-centric Behavioral Biometrics authentication system, utilizing unsupervised learning to identify users based on typing patterns without storing sensitive text data.
• Engineered a Multi-Task Learning pipeline using Random Forest and Support Vector Machine (SVM) classifiers to distinguish individual user signatures with high precision.
• Researched & Implemented advanced Anomaly Detection algorithms, benchmarking the performance of Cortical Learning (HTM) against Robust Random Cut Forest (RRCF) for outlier detection.
• Developed a modular, scalable Command Line Interface (CLI) framework using Python OOP principles to streamline data processing and recognition tasks.

Jan 2019 – Apr 2019 · 3 mos

• Developed a computational engine for Formal Concept Analysis (FCA), translating abstract mathematical models into performant code for data mining and knowledge representation.
• Implemented complex lattice structures and Order Theory algorithms, enabling the efficient mapping of objects to attributes (binary relationships).
• Engineered core lattice operations including Supremum (Join) and Infimum (Meet) to traverse concept hierarchies.
• Designed the API for calculating Intension and Extension, allowing users to programmatically derive concepts from data contexts.

May 2018 – Jul 2018 · 2 mos · New Delhi Area, India

• Worked as Full Stack Android Developer, build an e-learning app prototype and assisted in the implementation of IOT apps.

May 2018 – May 2018 · 0 mo · Kharagpur Area, India

• Worked as a research intern working on Human-Computer Interaction, Implemented detection of eye blinking from live video input, leading to the user's stress analysis.