AI

Graph RAG for enterprise documents: the graph is the easy part
Extracting entities and relations into a graph is the cheap step. In an enterprise, Graph RAG succeeds or fails on source authority, entity resolution, permissions, updates and evaluation. Here is how we build it.

LLM evaluation and reward hacking: how AI agents game evals, and how to build ones that hold up
A reference page on LLM evaluation and its main failure mode, reward hacking. Definitions, documented examples, how to detect gaming, how to design evals that survive production, and a checklist for enterprise teams.

Recursive self-improvement (RSI): what it is, the evidence, and the 2027 question
Recursive self-improvement is AI making the next AI better, faster than people alone could. A reference page: definition, history, mechanisms, the 2026 evidence, attributed forecasts, the counterarguments, and what it means for teams that build on AI.

Mechanistic interpretability in 2026: how it works, where it breaks, and what it means for production AI
A technical deep dive into how researchers reverse-engineer LLMs: superposition, sparse autoencoders and transcoders, attribution graphs, steering and the limits of chain-of-thought faithfulness. Plus how we use interpretability when shipping production AI agents.

AI automation companies in the UAE: a vendor index
Nine companies doing AI and automation work in the UAE, from Abu Dhabi AI groups to global consultancies and RPA platforms. Each entry says what kind of firm it is, where it sits and what it does, with a source you can check. Not a ranking.

Legacy system modernization with AI: replace, integrate or automate the screens?
Four ways to modernize a legacy ERP or portal with AI: replace it, integrate through an API, script it with RPA, or put an AI agent on the screens your team already uses. When each one fits, what to ask, and where to start.

AI consultant vs AI agency vs in-house: an honest comparison
A consultant helps you decide, an agency builds what you specify, an in-house team owns it for the long run, and a fixed-price pilot gives you numbers before you commit. Here is how they compare, and when you should not hire us.

AI consulting for small business: what to buy first, and what to skip
Most small businesses don't need an AI strategy. They need one task that eats the week done faster, on their real data, with a clear cost. Here is what to try yourself first, when outside help pays, what it should include and what to skip.

AI implementation consultant: what to hire for, and how to tell who can ship
An AI implementation consultant gets AI working inside your real systems, on your real data, and keeps it working. Hire for testing on your cases, integration, data handling, model choice, review design, monitoring and handover. Here is what each means and the questions that reveal who has shipped.
Older posts
- 11 Questions you should ask before building your own chatbot in 2019. AI
- Definitive Guide on Migrating from Dialogflow to Rasa AI
- Getting your Dataset ready for Machine Learning AI
- 5 Lessons we learnt from our large-scale enterprise AI implementation AI
- 5 Tips to use DialogFlow the right way AI
- Using EmoDet for Emotion Classification in Text AI
- How to teach Neural Networks to detect everyday sounds AI
- Realtime Object and Face Detection in Android using Tensorflow Object Detection API AI
- Why every TensorFlow developer should know about TFRecord! AI
- Using Artificial Intelligence to improve claims processing AI
- Here''s a very crucial thing people who build AI know, but you don''t AI
- How to easily extract Text from anything using spaCy AI
- Using AI to detect Facial Landmarks for improved accuracy AI
- Apache Spark - Which language to pick? AI
- I was bored and I made a bot AI
- Writing CNN from Scratch AI
- Train your Deep Learning models on the Cloud AI
- Deep Learning 101: How we design a Deep Learning Solution AI
- Leveraging on transfer learning for image classification using Keras AI
- Build your own Google Assistant Bot for your Firebase project Today AI
- Tensorflow Resources everyone should know AI
- A.I. Series Part 1 - Normalizing Data AI
- Building your own recommendation engine AI