Riddhi Gohil

Software Engineer

Datametica Solutions Private Limited•  September 2021 - August 2023

● Engineered and fine-tuned an SVM model for rumour detection, achieving a prediction accuracy of 92%, and demonstrating expertise in machine learning and data analysis. ● Spearheaded the development of data pipelines with Python, Apache Spark, Kafka, and Hadoop for real-time RSS feed ingestion into HDFS. Achieved exceptional efficiency, executing thousands of transactions in under 3 minutes, enhancing data processing speed. ● Employed selenium for scraping from live websites, processed and stored data in PostgreSQL after normalization and outlier detection. ● Formulated a robust pipeline with GCP APIs to collect videos, converting them to MP3 via ffmpeg and utilized speech-to-text libraries to extract speech content, achieving an accuracy improvement of 70% through segmenting large audio files. ● Implemented Google's Audio-to-Text model and transcription APIs for precise transcription and language translation, enhancing the accessibility of extracted information. Achieved seamless processing of multimedia data and cross-lingual content accessibility. ● Led the successful migration of a critical data processing pipeline from Alteryx to GCP's Data Fusion and Data Prep. This strategic move not only reduced operational costs significantly but also ensured seamless data flow within the GCP ecosystem.

Northeastern University, Boston

Computer Science, MS•  September 2023 - Present

Relevant Coursework: Distributed Systems, Programming Design Paradigm Using Java, Software Engineering, Database Management System, Cloud Computing, AI/ML