-
AI-Assisted Code Review: Advantages, Limitations, and the Role of a Developer
Currently, when discussing the topic of agentic engineering and, in particular, AI-assisted code review, people generally fall into two camps. Some argue that reading every change made by AI agents slows down development. Others are unwilling to…
-
Close Enough Isn’t Good Enough: Making AI Agents Build UI That Matches the Design
If your team uses AI agents for frontend tasks, this scenario may be familiar: the designer creates a page in Figma and the developer takes a few screenshots to feed them to an AI coding agent. The…
-
RAG vs. Fine-Tuning vs. Long Context: Choosing the Right Approach for LLM AppsÂ
Large language models are now widely used in software applications. However, relying on a model alone is not always sufficient to achieve the desired outcomes. Different tasks have different requirements, such as access to relevant information, consistent…
-
Running LLM Code Review in isolated AWS microVMs
Automated code reviews can run in isolated environments without relying on resources running 24/7. This article explains how to use disposable Firecracker microVMs on AWS to execute an LLM-based code review. It describes a small reference implementation…
-
How to Reduce Token Usage in AI Apps: A Healthcare App Use Case
As an AI-powered application grows scales, repeated AI requests often significantly increase AI requests and slow down response time, affecting user experience. This is where caching comes in handy. Instead of performing the same expensive operation every…
-
When Anthropic Prompt Caching Doesn’t Pay Off: Finding the Right Context Window
Extending the conversation history just to reach Anthropic’s minimum prompt caching threshold can cost more than keeping a shorter context window without caching. That’s what we found when optimizing a voice assistant whose responses rarely required more…
-
Wiring an AI Image Analyzer API Into Your App
Adding an AI image analysis feature to an app can be surprisingly easy: send an image to an API, as for the information you need, and process the response. The harder part is to make it reliable…
-
Why Vibe-Coded Apps Still Need Developer Review
A short answer: Vibe-coded apps can work well for prototyping and early validation, but they aren’t production-ready out of the box. Before release, applications built with AI still require a thorough engineering review to spot architecture, security,…
-
How We Handle Regression Testing In Large Applications
This article explains the hybrid regression testing approach our QA team uses on large projects, determining what needs to be tested and how thoroughly. The approach combines impact analysis and risk assessment to define the regression scope…
-
Evidence-First RAG: Validating LLM Citations in Application Code
This article explains how an evidence-first RAG design keeps citation validation in the application layer, ensuring that generated citations can be traced back to the evidence selected before generation. Key Takeaways This article is the final part…
-
Vector-Only vs Hybrid Retrieval for Science and Technology Books
This article examines when vector-only retrieval is enough for science and technology books, and when useful evidence can be missed. RAGLibrarian adds complementary retrieval methods, fuses their ranked candidates with hybrid RRF, and benchmarks whether extra retrieval…
-
How to Improve RAG Retrieval with Chapter-Aware Chunking in Go
The article explains how chapter-aware chunking can improve retrieval quality in RAG systems that use science and technology books as source material. In particular, it describes the implementation of RAGLibrarian – a custom Go-based page window chunker…
Boost your business with custom software
Tell us about your business needs and we’ll suggest a solution
Thank you!
We have received your request and will get back to you within 1 business day.











