Insights from the Consultancy

Notes from the Global Azure Bootcamp — API governance for AI workloads, the new shape of phishing, knowledge graphs as a context layer for LLMs, and why governance keeps showing up in every talk.

Notes from a meetup talk on real-world API breaches — the patterns that keep producing them, and the small, boring controls that prevent most of the damage.

Notes from building a RAG system over technical PDFs at D&V Electronics — the modality gap between text questions and image-only answers, and the practical fixes that closed most of it.

Notes from building CryptoRabbit.net — a digital asset marketplace with React Native, Next.js, and AWS Lambda, plus AI-powered listing checks that quietly do most of the moderation work.

A comparison of two systems for keeping financial truth — the 700-year-old double-entry ledger and the modern smart contract. Different trust models, different failure modes, different reasons to exist. Why most real systems end up using both, and how to decide which responsibilities belong on which side.

How CryptoRabbit stitches an on-chain escrow contract into a working marketplace — the indexer, wallet flows, shipping integrations, and the boundary between what belongs on-chain and what belongs in a boring backend.

A design retrospective on the smart-contract escrow that underpins CryptoRabbit — why a non-custodial escrow beat a platform wallet, how the state machine and timeouts are shaped, and the chain trade-offs between Ethereum (EVM) and Solana.

Notes from building PetPortal.ai — an AI advisor that promises pet-health answers in 60 seconds. Lessons on shipping applied AI alongside subscriptions, a community feed, and a workflow-automation backbone.

Most teams reach for "an AI" without picking which kind. There are really four useful patterns — vector search, retrieval-augmented generation, multi-agent systems, and training your own model — each with very different cost, complexity, and payoff. This article walks through when each one is the right tool, with examples from real systems including the pet portal product search.

Building multi-agent systems teaches the value of balancing complexity with efficiency. Simplicity, effective use of natural language, and clear examples are key to unlocking their full potential. This article shares practical lessons and highlights the role of natural language and examples in guiding agents effectively.

A practical mental model for building tools and applications with LLMs. Treat agents less like APIs and more like motivated junior collaborators — give them context, constrain the problem, and design tight feedback loops. Lessons from working alongside an AI-native startup founder on real products.

Infrastructure management used to be complex and time-consuming, often requiring deep technical knowledge and hours of troubleshooting. With AI, it’s now much easier and faster. This article outlines a three-step process for leveraging AI in infrastructure: setting up resources, debugging with AI digests, and iterating for optimization. This approach has made managing infrastructure more accessible, empowering more people to build efficiently and focus on strategic goals.

Managing asynchronous API states like `isBusy`, `isDone`, or `isError` can be challenging. Here’s a structured approach using NgRx to make it easier.

A Practical Guide to Structuring Smart and Presentation Components in Frontend Development

A technical breakdown of implementing auto-save capabilities similar to Google Sheets, focusing on data synchronization and efficient real-time data handling in a front-end application.

This article focuses on WebStorm which is an IDE created by the wonderful folks at JetBrains. The IDE provides great functional tools for web development such as git integration which will be discussed in this article. This article assumes the reader is familiar with git concepts including forking and rebasing.