About Me

Driving innovation and productivity working through various tech stacks and codebases. Here's a glimpse into my journey.

Lishash Music

  • Built MongoDB Aggregation Pipelines, REST APIs and EDA Reports for User Activity and Session Analytics to improve engagement, retention, and acquisition of approx. 5K+ downloads.
  • Developed Data Retrieval Pipelines for Graph Databases over natural language semantics using Neo4J Cypher and created logic-based functions for Auto Tagging and Relationship Mapping for sub-genre nodes.
  • Created multi-layered NLP Models for Extracting Audio Attributes, Vocal Separation, Language Translation, and an Audio Analytics Matrix to generate clustered datastores for the Recommender Algorithm, helping around 1.7K+ active users discover music and engage in jamming sessions with people of similar tastes.

HCL Technologies

  • Developed and optimized classical machine learning algorithms with the power of Quantum Computing to solve real-world problems efficiently.
  • Improvised Classical Algorithms for solving Real-World Machine Learning Problems using Quantum Computing, leveraging quantum parallelism and entanglement to enhance computational efficiency.
  • Envisioned and documented industry use-case workflows for Quanvolutional Neural Networks (QNNs) and Q-Fourier Transforms, building Proof of Concepts (POCs) to demonstrate the superiority of QNNs over traditional Neural Networks.
  • Benchmarked performance improvements, where a 20-qubit quantum machine was found to be approximately 52,429 times faster than classical counterparts in specific workloads.

Samsung PRISM

  • Developed a Real-Time Video Deblurring Framework leveraging advanced Neural Network techniques to enhance video clarity and reduce motion blur.
  • Applied various Neural Network architectures to design and implement an efficient real-time Video Deblurring Framework.
  • Generated a dataset of 10K+ image vectors using diverse data augmentation and masking techniques to simulate real-world blurring conditions.
  • Collaborated with faculty members to conduct additional research on continual learning approaches for improving model adaptability and efficiency.

Siemens

  • Designed and implemented a scalable Knowledge Graph framework by integrating open-source APIs, building data pipelines, and leveraging NLP for intelligent search and recommendation systems.
  • Integrated Open-Source APIs for data ingestion and developed aggregation pipelines to stream structured data into a Graph Database, generating Knowledge Graph networks with 1500+ artifacts and their metadata relationships.
  • Conducted research on Word Ontology and Semantic Similarity, designing a Text Embedding Framework to enhance the architecture's contextual understanding.
  • Engineered an Internal Search Engine backbone by implementing Topic Modeling, a Recommender System, and Automated Tag Generation using multi-layered NLP techniques.

Bajaj Finserv

  • Designed and developed scalable data pipelines, an in-house Data Lakehouse, and real-time analytics platforms to drive data-driven decision-making across multiple business functions.
  • Built ETL Data Pipelines for real-time event logs, incorporating cloud ingestion frameworks and Apache Table Formats to ensure efficient data processing.
  • Architected and developed an in-house Data Lakehouse system on the AWS PaaS ecosystem, enabling unified data storage and analytics.
  • Developed an End-to-End Customer Data Platform to power marketing campaigns and re-targeting strategies, ensuring communication governance and user journey analysis reports for a retail user base of approximately 75.8 lakh existing customers, real-time leads, and third-party prospects.
  • Created an Automated Monthly HEART Metrics Report to track product engagement statistics across 47+ metrics, optimizing user retention and performance enhancement.
  • Streamlined real-time health and KPI dashboards with 60+ panels by setting up Kafka streams from loggers and implementing database batching for efficient data visualization.
  • Collaborated on scaling a Lead Management System, handling a daily influx of 22,000+ organic leads, cross-subsidiary customers, and external prospects.
  • Contributed to the Innovation Team by developing AI/ML Proof of Concepts (POCs), including FAQ Chatbots, Bajaj Broking Lens, Mutual Funds Recommender Models, Data Unification Models, and Telegram Channel Analytics.