Minimum qualifications:

• Bachelor's degree in Computer Science, Artificial Intelligence, Machine Learning, Data Science, Engineering, or a related field, or equivalent practical experience.
• 5+ years of software engineering experience with Python.
• Hands-on experience building with LLM APIs, AI assistants, RAG pipelines, evaluation systems, model adaptation workflows, or AI-powered product features.
• Strong understanding of APIs, backend services, JSON, databases, authentication, permissions, and cloud-based application development.
• Comfortable writing clean, testable code, reviewing architecture, debugging production issues, and working with Git-based development workflows.
• Ability to move from prototype to production while thinking about reliability, security, observability, user experience, and operational cost.
• Curiosity for emerging AI systems and the engineering discipline to turn AI experiments into dependable product capabilities.

Job Description:

As an Applied AI Engineer at Aiotrix, you will turn modern AI capabilities into dependable production systems. This role focuses on model behavior, prompt optimization, RAG and context retrieval, evaluation pipelines, fine-tuning workflows, and measurable AI quality.

You will work on the intelligence layer behind ART and Aiotrix products: selecting model strategies, improving outputs, grounding responses in the right context, evaluating regressions, and making AI systems reliable enough for real users.

Responsibilities:

• Design and implement LLM-powered capabilities with strong focus on prompt optimization, structured outputs, grounding, model selection, and quality evaluation.
• Build retrieval-augmented generation pipelines using embeddings, vector databases, chunking strategies, reranking, metadata filtering, and relevance evaluation.
• Create context retrieval and memory strategies, including session context, task history, document grounding, vector retrieval, and context compression.
• Design evaluation pipelines for AI systems, including golden datasets, test cases, trace analysis, regression checks, hallucination checks, task completion metrics, and quality scoring.
• Develop fine-tuning or model adaptation workflows where appropriate, including dataset preparation, labeling strategy, experiment tracking, and performance comparison.
• Improve AI reliability through validation layers, output schemas, prompt tests, fallback strategies, guardrails, and human review loops.
• Analyze production AI behavior using logs, traces, user feedback, latency, cost, errors, and output quality metrics.
• Collaborate with backend, product, workflow, and systems engineers to integrate applied AI capabilities into production applications.
• Document model decisions, prompts, retrieval strategy, evaluation methodology, known limitations, and deployment behavior clearly.

Preferred qualifications:

• Experience building production software with Python, backend services, APIs, and cloud-based systems.
• Hands-on experience with LLM APIs, structured outputs, RAG, prompt design, model evaluation, and production AI behavior analysis.
• Experience with LangChain, LlamaIndex, retrieval systems, evaluation frameworks, experiment tracking, or model adaptation workflows.
• Understanding of embeddings, chunking, reranking, metadata filtering, semantic search, context windows, and retrieval quality trade-offs.
• Experience with vector databases or retrieval systems such as pgvector, Qdrant, Pinecone, Weaviate, Milvus, Chroma, Elasticsearch, or similar tools.
• Familiarity with AI evaluation, hallucination testing, output validation, prompt regression, and production monitoring.
• Ability to reason about model reliability, retrieval quality, grounding, data quality, latency, cost, and safe AI behavior.
• Strong product thinking and the ability to convert model capabilities into practical AI features for users and teams.