Join the Kreitech Team!

We're looking for an Applied AI Engineer to join our team and build production-grade agentic systems for clients in fintech and regulated industries. If you've gone beyond demos and know what it takes to ship agent systems that hold up in month six — reliable, observable, and improvable from their own production data — we want to hear from you.

Responsibilities:

- Design and build production AI agent systems embedded directly in client environments: CRM automation, document processing, compliance workflows, and enterprise integrations
- Develop native MCP servers, sub-agents, and agent skills that connect LLMs to real client systems (APIs, data stores, enterprise platforms)
- Implement evaluation pipelines and regression suites — including LLM-as-judge grading and production replay — to maintain quality after deployment
- Instrument agent workflows for full observability: trace every decision, tool call, failure, and human correction in context
- Attribute production failures to the correct layer (prompt, context, tool, routing, or workflow design) and ship targeted, verified fixes
- Codify every client engagement into a reusable deployment pattern that the next engineer can run without the original author
- Run technical discovery with client stakeholders to understand how work actually happens — including the undocumented parts — and translate that into executable SOPs for the agent
- Own the human-review design: where a person stays in the loop, why, and what they're checking

Requirements:

- 4+ years in software engineering, technical consulting,



or a client-facing engineering role
- Production experience with LLM application development: prompt engineering, agent frameworks (LangGraph, LangChain, or equivalent), and RAG pipeline design
- Python as primary working language; SQL fluent; TypeScript a plus
- Experience building MCP servers, tool-calling agents, and stateful multi-step agentic workflows
- Comfortable integrating with enterprise systems via REST APIs, webhooks, and standard authentication patterns
- Can attribute a production agent failure to its root cause — instruction, context, tool, routing, or workflow design — in writing, unaided
- Strong written and spoken English for client communication and technical documentation

Desirable:

- Background in fintech, healthcare, insurance, or other regulated industries
- Experience with AI observability and evaluation tooling: LangSmith, Langfuse, LLM-as-judge, structured production replay
- Familiarity with enterprise client stacks: Azure, MS365, SharePoint, HubSpot, Power BI
- Prior consulting, professional services, or forward-deployed engineering experience
- Exposure to multi-agent coordination patterns and MCP ecosystem tools

Why join Kreitech?

- Remote work – enjoy the flexibility to work from anywhere
- Adaptable schedule – prioritize work-life balance
- Continuous learning opportunities – stay ahead in your field
- Competitive compensation in USD

Apply now for the Applied AI Engineer position at Kreitech:

https://airtable.com/appkPBsD5JS0Tp5wx/pagi7lO7hibX0U4bN?prefill_Position=Applied%20AI%20Engineer

📌 Applied AI Engineer (Argentina)
🏢 Kreitech
📍 Argentina