Ashwani Thakur

Currently leading software development initiatives at Spinny, focusing on architecting scalable systems and driving significant tech and operational efficiencies.

• Achieved fast-tracked promotion from SDE-II to Lead Software Engineer in 2024, recognized for exceptional performance in delivering high-impact solutions.
• Architected an event-driven refurbishment system leveraging Kafka, Redis, and PostgreSQL, which reduced vehicle turnaround time by 32% across thousands of units.
• Led the successful launch of Spinny's Insurance platform, designing a scalable, plug-and-play backend broker integration system.

Contributed as a Software Development Engineer II at Spinny, optimizing API performance and developing critical systems to enhance user experience and operational flow.

• Optimized multiple critical APIs, achieving a 10x reduction in latency to deliver faster response times and a smoother user experience.
• Architected a fault-tolerant, scalable car reservation and notification system using Kafka, Redis, and MySQL, increasing test drive flow by approximately 10% through asynchronous task processing and event-driven notifications.
• Designed a scalable 'Saved Search' system supporting over 1M filters with Elasticsearch, Redis, and Kafka for real-time matching and 7-day counts, optimizing storage via encoded keys.
• Led the development of multiple major platform features, including car exchange and budget car functionalities, significantly enhancing user engagement and profitability.
• Personalized user listing and homepage experiences, leading to improved recommendations and enhanced user engagement.

As a Software Development Engineer I at Spinny, developed and optimized microservices and rule engines to improve customer support efficiency.

• Developed the Helpcenter microservice, streamlining help and support queries to significantly enhance customer satisfaction.
• Designed and implemented a rule engine for ticket resolution, reducing inbound and outbound support calls by 30% and improving overall operational efficiency.

Developed robust backend systems for a scalable, serverless data management and marketplace platform at CrossML.

• Developed a scalable, serverless backend for a data management and marketplace platform utilizing AWS Lambda, Python, and DynamoDB.
• Designed highly optimized RESTful APIs with efficient caching and event-driven patterns to ensure seamless cross-service operations.

Contributed to Life.ai by designing and optimizing the graph database schema for their recommendation engine.

• Designed the Neo4j graph database schema for the recommendation engine, optimizing queries for high-performance recommendations.

Applied machine learning and computer vision techniques as a Data Science Intern at CrowdANALYTIX to develop product classification and predictive models.

• Built CNN-based product classification filters specifically for e-commerce images, enhancing categorization accuracy.
• Developed a predictive Machine Learning model to analyze and forecast trends within mortgage lending data.
• Extracted text from product label images using YOLO and Tesseract, improving data capture and processing capabilities.

Engineered a fault-tolerant circuit breaker package for Python web frameworks (FastAPI, Django) with in-memory and Redis storage options, state change alerts, and a fallback mechanism to enhance microservice resilience.

Research paper accepted in the International Conference on Signal Processing and Communication (ICSC 2019), co-authored with Dr. Krishan Kumar Saluja and Savita Walia.