AI integrator
| Highest | $7,350 / mo · Switzerland |
|---|---|
| Experience | 2–4 yrs |
| Degree | Not required |
| First paycheck | ~2 years |
Pay in the destination countries
Monthly, in US dollars. Each country is an Imora admission package — open one for that route's fees, language requirement and deadline.
| Country | Monthly |
|---|---|
| SwitzerlandIELTS 6.5 · CHF 0.9–8k | $7,350max |
| United StatesTOEFL 80–100 · $10–70k | $7,000 |
| NorwayIELTS 6.0–6.5 · NOK 130–260k | $5,600 |
| DenmarkIELTS 6.5 · DKK 45–135k | $4,750 |
| IrelandIELTS 6.0–6.5 · €10–25k | $4,600 |
| Hong KongIELTS 6.0–6.5 · HK–200k | $4,350 |
What does the day look like?
You build RAG pipelines: extract text from PDF, DOCX and scanned images with Unstructured.io, PyMuPDF and LlamaParse, split it into chunks, embed it with OpenAI text-embedding-3, Cohere Embed, Voyage AI or BAAI/bge, and index it in Pinecone, Qdrant, Milvus, Weaviate or PostgreSQL with pgvector. Then you write backend microservices in Python or TypeScript with FastAPI, LangChain, LlamaIndex, Haystack, Semantic Kernel and DSPy, wire tool-calling to Salesforce, HubSpot, SAP, Slack and Telegram, add safety rails with Guardrails AI and NeMo Guardrails, track latency, token use and errors in Langfuse, Arize Phoenix and Helicone, pack it into Docker containers and ship through CI/CD.
The imagined day is 80% chatting with models. The real split is 70% to 80% data engineering, corrupt PDFs, schemas, HTTP timeouts, rate limits and unit tests, 15% to 20% prompts, evaluation and guardrails, and 5% to 10% sprints, tickets and docs. The week is 38 to 40 hours, 37.5 hours in the United Kingdom, with almost no shift work, but production on-call is standard: 15 to 30 minutes to respond if an agent fails or a provider goes down.
Where does this lead?
Next step up
People arrive from
What do you make of this job?