-
Continue reading →: A Practical ECS Architecture for Building RAG SystemsWhen I started moving toward the role of an AI Integrator, I experimented with many “typical” AI project patterns and documented the process in my blog. One of the most revealing steps was building a RAG system. As a backend engineer, my instinct was to implement everything on AWS Lambda,…
-
Continue reading →: Explain me transformer as if I was 5For a long time, I couldn’t grasp what “attention” really was. Most explanations just threw formulas at me or repeated vague phrases like “compute query, key, and value with a feedforward net.” But I wanted something deeper. Not just what happens — but why it works.
-
Continue reading →: How I Automated Meeting Notes with LLM and Confluence IntegrationThis article covers automating meeting minutes with AWS, from transcription to summarization using LLM, and publishing to Confluence. It discusses challenges with large files and LLM formatting, along with the project architecture and Serverless deployment.
-
Continue reading →: From Feedback to Insights: Sentiment Analysis with Athena and QuickSightWe successfully completed the project by extracting feedback from the S3 bucket, analyzing sentiment in DynamoDB, and visualizing the data in QuickSight using Athena. This streamlined the process of analyzing sentiment trends, providing valuable insights for continuous product and service improvement.
-
Continue reading →: Prototyping Insights from Customer Feedback with PandasHere are example graphs for two categories, showcasing the data insights before proceeding with more advanced visualizations in QuickSight. These visualizations help us understand trends and patterns, providing a clearer foundation for building more comprehensive analysis tools for decision-making.
-
Continue reading →: High-Performance Sentiment Analysis Pipeline with SQS and LambdaRunning a large business with high volumes of feedback? Addressing each entry individually isn’t feasible. Here’s a serverless solution using Lambda and SQS to process feedback at scale. This architecture ensures efficient, cost-effective handling of large feedback volumes, saving time and resources while maintaining high performance.
-
Continue reading →: Building a Sentiment Analysis Pipeline on AWSThis post covers the first part of the implementation, orchestrating the workflow with AWS Step Functions and Lambda. We use DynamoDB for storing records, Amazon Comprehend for sentiment analysis, and the Serverless Framework with Python for provisioning Lambda functions, all within the AWS ecosystem.
-
Continue reading →: AWS AI Practitioner Certification: My Journey & Prep TipsIn this article, I share my experience with the AWS AI Practitioner exam, preparation tips, and resources I found effective. I also discuss emerging career opportunities in AI and cloud technologies, and how a structured study plan helped me balance learning with work.
