Skip to main content
Every Spark digest is built around the categories you select. Pick at least one and at most three. Three is the maximum because more categories dilute the daily ranking and the digest becomes noisy. Categories are grouped into three families and you can mix freely across families.

​
The eight categories

​
Technology

  • AI & Machine Learning. Artificial intelligence, machine learning models, LLMs, and real-world AI applications.
  • Developer Tools. Frameworks, CLIs, IDEs, open-source projects, and tooling that improve developer productivity.
  • Data & Analytics. Data engineering, analytics workflows, databases, and practical insights from metrics.
  • Knowledge Graphs. Knowledge graphs, ontologies, taxonomies, semantic web, linked data, graph databases, entity resolution, and knowledge representation.

​
Events & Marketing

  • Events & Community. Event technology, conferences, exhibitions, trade shows, community building, community-led growth, hybrid events, attendee experience, and event platforms.
  • AI & Marketing. AI tools for marketers, AI agents, marketing automation, generative AI for content, personalization engines, MarTech, and AI-powered analytics and attribution.

​
Business & Growth

  • Startups & Business. Startup execution, product strategy, developer careers, hiring, and the business of technology.
  • Creator & Growth. Creator economy, community monetization, audience growth, no-code automation, content strategy, newsletter growth, and event marketing ROI.

​
Pick your first set

During onboarding, Spark analyzes the website URL you provide and pre-selects up to three categories that match. Review the picks, add or remove as needed, then continue. If you skip the website analysis or want to start over, the picker opens with no categories selected. Pick at least one. The digest cannot run without an active category.
One to three is the hard limit. The picker disables the add button once you reach three and disables the remove button once you reach one. This is intentional: the digest ranker compares stories across your active categories, and too many categories produce a thin, repetitive feed.

​
Change categories later

Open the News page, then Manage sources. The same picker opens. Add or remove categories and save. The next collection cycle uses the updated set, and the change reflects in the next morning’s digest. Sources are derived from your categories automatically. Each category brings a curated bundle of trusted publication sources (industry blogs, subreddits, arXiv categories, news sites) so you don’t have to manage source URLs yourself.