How to hire a
AI engineer.
AI engineers are the scarcest senior role on the market in 2026. Hiring well requires a focused brief, ruthless evaluation of evals literacy, and patient compensation conversations.
Hire in-house
You've shipped one AI feature, you have ongoing AI roadmap, and the work justifies $200k+/year for several years.
Outsource to a studio
You're shipping the first AI feature, you don't have evals or RAG experience, and you need the build in 8-12 weeks.
Signs of strong candidates.
Strong signals
- Has shipped a production AI feature with evals (not just demos)
- Talks fluently about hallucinations, hybrid search, reranking
- Has opinions on RAG vs fine-tuning trade-offs
- Knows current frontier model strengths and weaknesses
- Has read recent papers and can summarize takeaways
Red flags
- ✗Confuses RAG with fine-tuning
- ✗Has built only chatbots with no evals
- ✗Resume-driven LLM name-dropping without depth
- ✗No production deployment experience
- ✗Vague answers on cost or latency
Questions to ask.
- Q1
Walk me through an AI feature you shipped to production. What did the eval set look like? What was your worst regression?
- Q2
When would you fine-tune vs RAG vs prompt-engineer?
- Q3
How do you handle prompt injection in retrieved content?
- Q4
How would you evaluate a new RAG system before launch?
- Q5
What's the cheapest way to ship a chat feature serving 100k queries/day?
We bridge
to your hire.
We help clients write the brief and run technical screens. If you're not ready to hire a AI engineer yet, brief us — we can fill the gap and help you hire later.
Brief us