What is Ninjagraph?
Ninjagraph is Ninjahire's internal candidate database. It's the source the platform searches through by default when you run AI sourcing for a job and haven't specifically selected other channels — in other words, if no other sourcing channels are chosen, NinjaGraph is what powers your search.
That said, Ninjagraph is the primary source Ninjahire searches through when no other channels are selected. Other sources, like job boards (Indeed, Monster, Dice, etc.) and your ATS, are additional channels you can layer on top of Ninjagraph, not replacements for it. So by default, your search runs against Ninjagraph, and you can expand outward from there by selecting additional channels alongside it.
How does it work?
Ninjagraph's search isn't just keyword matching — it understands the meaning behind what you're looking for. Specifically, it interprets:
Skills
Seniority
Similar or related job titles
Industries
Technologies
Years of experience
Semantic meaning — the intent behind a phrase, not just the literal words
This means you can search in plain English and get relevant results. For example, if you search:
"Fullstack developer with 4 years of experience in React and Node JS for Austin location"
Ninjagraph will return matching profiles based on that full description — recognizing "Fullstack developer" as a role, "4 years" as an experience filter, "React and Node JS" as required skills, and "Austin" as a location filter — rather than needing you to search each of those as separate, rigid fields.
What is the AI actually doing?
Behind the scenes, when you run a search or AI sourcing, the AI performs a few specific tasks to turn your query into a ranked candidate list:
Understanding synonyms — recognizing that different terms refer to the same skill, role, or concept (so you don't miss candidates just because their profile uses different wording than your search).
Extracting skills — pulling out the specific skills mentioned in your search (or in the job description) to match against candidate profiles.
Normalizing titles — mapping different variations of a job title to a consistent standard, so "Full Stack Engineer," "Fullstack Developer," and "Full-Stack Software Engineer" are all recognized as the same role.
Ranking candidates — scoring and ordering results based on how closely each profile matches the job title and search criteria, so the strongest matches surface first.
To summarise Ninjagraph is the default internal database Ninjahire searches candidates from. You can expand your search to additional channels, and the AI behind the search understands plain-English queries — matching on skills, seniority, titles, industries, and experience — then ranks the results so the best-fit candidates appear first.