Chiang Rai Think Tank

The Promptsmith

I know what makes prompts fail — and how to make them bulletproof

The Promptsmith knows every major LLM inside out — GPT-4, Claude, Gemini, Llama, and beyond. He crafts, refines, and stress-tests prompts so...

The Architect

Hard technical decisions, made with confidence — and the WHY behind every choice

The Architect makes hard technical decisions with confidence. From model selection to multi-agent systems, he explains the WHY behind every ...

The Evaluator

If you can't measure it, you can't trust it — LLM evaluation that actually works

The Evaluator builds benchmarks, designs eval pipelines, and finds where your LLM actually fails. No vanity metrics — just honest measuremen...

The Pipeline Builder

From prompt to production — RAG, agents, and pipelines that ship

The Pipeline Builder turns AI ideas into working systems. RAG pipelines, agentic architectures, API integrations — he builds the plumbing th...

The Cost Mind

Stop wasting money on AI — cost models that actually add up

The Cost Mind models AI spend, optimizes token usage, and compares providers so you stop guessing and start knowing exactly what your AI cos...

The Product Strategist

AI products people actually want — strategy, positioning, pricing

The Product Strategist turns AI capabilities into products people pay for. Positioning, pricing, go-to-market — he builds the strategy that ...

The Data Curator

Garbage in, garbage out — training data that actually makes your model better

The Data Curator builds training datasets, designs eval sets, and plans fine-tuning strategies. Because the best model in the world is usele...

The Red Teamer

I find what breaks before your users do — adversarial AI testing

The Red Teamer probes for jailbreaks, maps failure modes, and stress-tests guardrails. He doesn't pull punches — his value is finding the cr...