استخدام Senior Applied AI Engineer

Role mission

You will help build the "brain" of Mayia: the systems responsible for extracting, structuring, connecting, retrieving, and preparing knowledge for users and AI agents.

This role is suitable for someone who understands LLMs and RAG but is equally interested in software engineering, knowledge representation, graph-based systems, evaluation, and production reliability.

Responsibilities

• Design and develop LLM-, embedding-, and knowledge-graph-based retrieval systems.
• Build pipelines for extracting entities, relationships, facts, and metadata from organizational data.
• Develop knowledge storage and retrieval capabilities using Graphiti, Neo4j, or similar technologies.
• Design entity-resolution, deduplication, and contradiction-detection mechanisms.
• Develop topic or matter detection and route information to the appropriate graph partitions.
• Handle temporal changes, fact invalidation, and replacement of outdated information.
• Develop working-memory and context-building capabilities for language models.
• Contribute to intent detection and understanding the user’s actual information need.
• Implement source attribution, confidence, trust, and explainability mechanisms.
• Build benchmarks, golden query sets, and regression tests for retrieval and generated answers.
• Improve the cost, speed, and quality of large-scale document processing.
• Work closely with backend and product engineers to expose AI capabilities through reliable APIs and user experiences.

 

Required qualifications:

•  Strong production software-development experience with Python.
•  Practical experience withLLM APIs, prompt engineering, embeddings, andRAG.
•  Experience with knowledge graphs, graph databases, or entity-linking systems.
•  Good understanding of information retrieval, semantic search, and ranking.
•  Ability to build maintainable data-processing pipelines and services.
•  Experience evaluating LLM systems and diagnosing retrieval failures or hallucinations.
•  Familiarity with Git, Docker, automated Testing, and collaborative development.
•  Professional written and spoken English.
•  Ability to convert research-oriented problems into practical engineering solutions.

 

Nice to have:

• Experience with Graphiti, Neo4j, or temporal knowledge graphs.
•  Experience building entity-resolution and data-deduplication systems.
•  Experience processing legal, enterprise, or sensitive documents.
•  Familiarity with agentic workflows and tool calling.
•  Research experience inNLP, machine learning, explainable AI, or knowledge representation.
•  Experience with asynchronous and parallel processing at scale.
•  Familiarity with clustering, topic detection, or representation learning.

 

Measures of success:

• During the first few months, you should be able to:
•  Measurably improve retrieval and source-attribution quality.
•  Establish a repeatable evaluation benchmark.
•  Reduce duplicate entities and contradictory information.
•  Improve the reliability of historical graph generation and incremental updates.
•  Convert experimental AI capabilities into maintainable product services.