Michelle Ten·Aug 13A more complex or sophisticated technique is not automatically the better choice.I FINALLY understood lesson we were supposed to learn from one module of a course I attended 6 months ago… by building an application.
Ram Chandra Kambalapally·Jul 26Building an AI Assistant That Actually Works: Part 1 — The 800-Millisecond MailroomWhat really happens in the fraction of a second between a user hitting “send” and an AI replying? Follow one message through nine desks and…
Sonal Mishra·May 22I Fine-Tuned DistilBERT Twice. It Humbled Me Both Times.Same model. Same three epochs. Weighted F1 of 0.9070 on one. Weighted F1 of 0.7261 on the other. Here’s why — and what the gap actually…
Sonal Mishra·May 22How I Fine-Tuned DistilBERT to Classify Complaints — And What I Learned Along the WayThe Setup: AI Wrote the Code — I Made Sure I Could Explain Every Line
Naoki Goto·Apr 10Implementing CI/CD Pipeline with GitHub ActionsStreamlining the deployment and eliminating human errors
Balamukunda D R·Apr 4How Machines Read Between the Lines — A Beginner’s Guide to NLP Token ClassificationUnderstanding the building blocks that power chatbots, search engines, and AI assistants
Balamukunda D R·Apr 4Fine-Tuning DistilBERT for POS Tagging & Chunking — What I LearnedA hands-on walkthrough of token classification using Hugging Face Transformers
Naoki Goto·Mar 27Deploying DistilBERT on AWS EKSFrom Minikube to AWS EKS: Infrastructure as Code, Load Testing, and What CPU Throttling Does to Inference Latency
