AI / Automation Engineer

We are looking for a Python engineer who builds AI-powered features and automated workflows. You will be comfortable moving between backend APIs, LLM integration, retrieval systems and workflow automation.

You will design and ship AI features end to end: data goes in, a model sits in the middle, and the business gets a reliable result it can depend on.

Experience: 2 to 5 years.

What you will do:

• Build and maintain Python backend services and APIs
• Integrate LLMs into product features, including prompting, structured output and tool calling
• Design and tune RAG pipelines: chunking, embeddings, vector search and retrieval quality
• Build automated workflows that connect systems, APIs and AI steps
• Work with vector and relational databases
• Measure and improve the accuracy, speed and cost of AI features
• Write tests and documentation alongside your code

What we look for:
We value curiosity and self-direction. This field moves fast, and we want someone who keeps up because they want to, not because they are told to. We also value honesty about what you do not know, a habit of measuring instead of assuming, and real care about whether a feature works in practice, not just whether it demos well.

Must have:

• 2 to 5 years of professional software development in Python
• Experience with FastAPI, Flask or Django, including REST API design, validation and async
• Hands-on experience integrating LLM APIs into real applications
• Working knowledge of RAG: embeddings, chunking and retrieval
• SQL and relational database design
• Git, code review and testing discipline
• Clear written communication

Strongly preferred, automation:

• n8n (or Make, Zapier, Airflow or Temporal) for building multi-step workflows
• API integration and webhook-driven automation
• Scheduled jobs, queues and background processing
• Automating internal and operational processes end to end

Strongly preferred, AI:

• Agentic RAG: query planning, iterative retrieval and self-correction
• AI agents: tool calling, multi-step reasoning and guardrails
• Vector databases such as Qdrant, pgvector, Chroma, Pinecone or Weaviate
• Hybrid search and reranking
• Prompt and context engineering
• Evaluating AI output: accuracy, grounding and hallucination control
• Running open-source models locally with tools like Ollama, vLLM or Hugging Face
• Frameworks such as LangChain, LlamaIndex or similar
• MCP (Model Context Protocol) and current agent tooling

Nice to have:

• Docker and basic Linux
• Document processing: OCR, PDF parsing and vision models
• Frontend familiarity (React or TypeScript)
• Cloud platforms (AWS, GCP or Azure)
• Open-source contributions or personal AI projects