Nyah Macklin

Nyah Macklin is a Senior Developer Advocate for AI at Neo4j and the founder and CTO of Afros in AI, a technical community exploring artificial intelligence. Their developer education ranges from helping beginners publish their first website to teaching context engineering: combining prompts, retrieval, and knowledge graphs so AI applications can use connected information and expose the sources behind their answers.

Making software accessible

Nyah entered software through Resilient Coders, a paid training program. Growing up without a laptop, they had understood technology as a field for people with money; free software and inexpensive development tools helped make a different career possible. Their earlier work included service as a chief of staff in Massachusetts government.

At Suborbital Software Systems, Nyah was the fourth engineering hire and helped build the company’s flagship products. They later became a developer evangelist at Couchbase and taught women and nonbinary engineers at G{Code}. Their account of adapting to technical change includes a revealing reversal: Nyah initially avoided React and Hooks because their model of state felt unfamiliar, then recognized that avoidance as an obstacle to learning. React eventually became central to their development experience. Their willingness to reconsider a tool comes with practical questions about its maintenance, cost, funding, and effects on users.

At Couchbase, Nyah worked with Jessica Rose on Learning on the Couch, helping beginners move from understanding and installing a code editor to version control and publishing a website with Netlify. The lessons explain the connections experienced developers can take for granted: how code written locally reaches GitHub, and how that code becomes a website another person can visit.

Nyah also addresses developers’ responsibilities toward the people whose data they handle. Their guidance on ethical data stewardship brings consent, data minimization, privacy expertise, and testing into development planning. A hypothetical health application makes those decisions concrete by showing how a team might organize and test privacy protections.

From graph fundamentals to connected AI context

At Neo4j, Nyah teaches developers how relationships between pieces of information change what an application can retrieve and explain. In 2025, Nyah and Martin O’Hanlon created an introductory graph-database video series covering nodes and relationships, data imports, and using Neo4j from Python. Their subsequent AI instruction develops that foundation in several directions:

• Retrieval beyond similarity: In RAG Needs a Map, Nyah combines vector search with a knowledge graph built from course material. Vector search finds semantically similar lessons; graph queries retrieve related information and support questions requiring traversal or aggregation. An extraction schema specifies entity and relationship types, guiding the LLM’s construction of the graph. This gives developers a structure they can inspect and query alongside the retrieved text. When an agent returns a lesson count, Nyah checks the underlying database query—a concrete example of examining how an answer was produced rather than accepting its presentation.

• Context across interactions: Nyah’s context-engineering curriculum connects prompt management, retrieval-augmented generation, and knowledge graphs, using Neo4j as a memory layer. The design question extends beyond finding a relevant document: developers must decide what information an application needs to retrieve and retain across interactions.

• Relationships and decision history: With Zaid Zaim and Alexander Erdl, Nyah co-authored an exploration of context graphs and AI memory. The article examines how explicit relationships and recorded decisions can make organizational knowledge queryable and auditable. It brings together multiple practitioners’ approaches, extending the discussion of agent memory beyond stored text to the connections and decision history an organization may need to recover.