Insights
Field notes on building AI that works
Long-form, practical writing from engineers who build and run AI systems. No hype, no listicles — the things we wish someone had told us.
Latest · 2 Nov 2026 · 6 min read
What Is Agentic AI Automation? A Practical Guide for Business Teams
Agentic AI automation explained without the hype — how AI agents differ from chatbots and RPA, where they work well, where they fail, and how to start safely.
-
26 Oct 2026 · 6 min
Designing Tools for AI Agents: Patterns That Make Agents Reliable
Most agent failures are tool-design failures. Practical patterns for naming, scoping, validating and securing tools so LLM agents choose and use them correctly.
- AI agents
- tool calling
- engineering
-
19 Oct 2026 · 6 min
The Production RAG Checklist: 25 Things to Get Right Before Launch
A practical checklist for building retrieval-augmented generation systems that work in production — ingestion, chunking, retrieval, grounding, permissions, evaluation and operations.
- RAG
- retrieval
- LLM
-
12 Oct 2026 · 6 min
LoRA vs QLoRA vs Full Fine-Tuning: Choosing the Right Method
How LoRA, QLoRA and full fine-tuning work, what they cost in memory and compute, how quality compares, and a decision guide for fine-tuning open-weight LLMs.
- fine-tuning
- LoRA
- QLoRA
-
5 Oct 2026 · 6 min
How to Prepare a Fine-Tuning Dataset That Actually Improves Your Model
A step-by-step guide to building instruction datasets for LLM fine-tuning — sourcing examples, labelling guidelines, formatting, deduplication, synthetic data, splits and quality checks.
- fine-tuning
- datasets
- data labelling
-
28 Sept 2026 · 6 min
How to Evaluate an LLM Application: A Practical Guide
How to build evaluations for LLM features, chatbots, RAG systems and agents — defining quality, building test sets, choosing metrics, calibrating LLM judges and running evals in CI.
- evaluation
- LLM
- testing
-
21 Sept 2026 · 6 min
Ollama vs vLLM: Choosing an Engine for Self-Hosted LLMs
A practical comparison of Ollama and vLLM for running open-weight LLMs on your own servers — ease of use, throughput, hardware, quantisation, APIs and production operations.
- self-hosted LLM
- Ollama
- vLLM
-
14 Sept 2026 · 6 min
Human-in-the-Loop Design for AI Automation: Approvals That Don't Slow You Down
How to design human oversight for AI agents and automations — risk tiers, approval thresholds, review UX, shadow mode, escalation and graduating to autonomy.
- AI agents
- human-in-the-loop
- governance
-
7 Sept 2026 · 6 min
Model Context Protocol (MCP) Explained for Business and Engineering Leaders
What the Model Context Protocol is, why it matters, how MCP servers work, real use cases, security considerations and how to decide whether your organisation or product needs one.
- MCP
- AI integration
- agents
Subscribe via RSS.
Start a project
Have a workflow that should run itself?
Tell us what you want automated or built. You get a written assessment of feasibility, risks and a fixed quote — usually within two business days.