Jun 2024 – Nov 2024 · 5 mos · Cupertino, California, United States · Remote

• Architected and implemented a comprehensive backend infrastructure using Golang (Gin Framework) and MySQL database.
• Features built: Payments Service, Job Posting and Application Flow (async video interviews, applicant search with custom filters, AI applicant evaluation), User Engagement Modules (Profile Timeline, Notifications, Vouching Mechanism, Checklists)
• Enhanced system performance by 70% through Redis caching, strategic MySQL database indexing, asynchronous workers and efficient pagination in REST APIs, improving response times and reducing server load.
• Built a Resume parser using Langchain and OpenAI in Python (FAST API Framework) which led to the onboarding of 20,000 users.
• Developed a search engine using Meilisearch with dynamic filters and faceted search for the recruiter-facing dashboard.
• Automated Open Graph Image generation using NodeJS and Vercel-Satori, optimizing content sharing across platforms.
• AWS Infrastructure Management:
• Orchestrated the migration of four services from AWS AppRunner to EC2 instances with auto-scaling and load balancers, improving system visibility and performance.
• Implemented Infrastructure as Code (IaaC) scripts through Pulumi and Bash for CI/CD deployments and scalable, cross-region, multi-environment replication.
• Built end-to-end HLS video streaming using AWS Elemental MediaConvert, Lambda, CloudFront, and S3.
• Implemented a CDN-based image-serving framework using AWS Cloudfront and S3.

Nov 2023 – May 2024 · 6 mos · Bangalore Urban, Karnataka, India · Hybrid

• Designed and implemented ETL pipelines using Apache Airflow and Mage AI with Python to
• efficiently process vast datasets originating from diverse food and retail platforms like Amazon, and Foodpanda, and social media platforms like Instagram and Twitter.
• Managed the integration of these pipelines on cloud servers like Jenkins to handle approximately a billion data points, ensuring accurate tagging of products/posts using Natural Language Processing (NLP) and language-specific tokenization tailored to specific client requirements.
• Implemented multithreading and parallel processing within these pipelines for enhanced efficiency, reducing the run time by 50%.
• Worked with AWS services such as EC2, S3, DynamoDB, and Athena for enhanced data management.
• Performed data collection from retail platforms using BeautifulSoup and proxies, and data collection from the Google Trends API to obtain Google Search and YouTube Search data of specific trends.
• Key technologies utilized include Python, AWS, REST APIs, ElasticSearch, Apache Airflow, Jenkins Mage AI, and Docker to streamline data processing and optimize workflow efficiency.