I Analyzed 4.78M HN Items to Create an Interactive 2025 Infographic

I downloaded the complete Hacker News archive (160 SQLite shards via StaticNews) and analyzed all 2025 data to create a GitHub Octoverse-style interactive infographic. https://andreyandrade.com/static/hn-universe/ The AI conversation on HN shifted dramatically: Programming Languages Zig is the most hyped language of 2025. Ruby had a surprising comeback. Job Market (4,885 postings analyzed) Remote: 57% ████████████████████░░░░░░░░░░░░░░░
Onsite: 23% ████████░░░░░░░░░░░░░░░░░░░░░░░░░░░
Hybrid: 20% ███████░░░░░░░░░░░░░░░░░░░░░░░░░░░░ Some surprising discoveries: The analyzers are written in Go and process all 160 shards concurrently. Each shard contains ~30k items. The trickiest part was normalizing the data - HN's API changed formats over the years, and some items have missing fields. What Surprised Me Most Tell HN vs Show HN: I expected Show HN (project launches) to get more engagement than Tell HN (personal stories). It's the opposite - Tell HN averages 54.7 points vs Show HN's 11.2. The community values personal narratives more than product launches. The DeepSeek Spike: In January 2025, DeepSeek dominated discussions. By March, mentions dropped 92%. The AI hype cycle moves fast. Junior Jobs Scarcity: Only 3% of "Who is Hiring" posts target juniors. The industry's entry barrier keeps rising. The infographic has smooth scroll animations and works on mobile. All data is from real HN submissions and comments. What patterns have you noticed on HN this year? Built by https://andreyandrade.com - Backend engineer focused on Go, data analysis, and AI integration. Templates let you quickly answer FAQs or store snippets for re-use. Are you sure you want to hide this comment? It will become hidden in your post, but will still be visible via the comment's permalink. Hide child comments as well For further actions, you may consider blocking this person and/or reporting abuse - AI-related stories grew from 8% to 13% of all submissions
- Claude overtook ChatGPT in mentions during H2
- DeepSeek was January's phenomenon, then vanished (-92%)
- The "Models Race" shows Claude at +482% growth vs GPT's +65% - 71% mention AI/ML skills
- Only 3% are junior positions
- Average salary: $180k (when disclosed) - "Tell HN" posts average 5x more points than "Show HN"
- Best posting time: 15:00 UTC (11am EST)
- September had the most viral posts (34 stories with >1000 points)
- Weekend posts get 23% less engagement - Data: https://static.hn (160 SQLite shards, 4.78M items)
- Analysis: Go (processes all shards in parallel)
- Visualization: Vanilla HTML/CSS/JS with CSS animations - Downloaded all 160 shards (~2GB total)
- Wrote Go scripts to query patterns across shards
- Aggregated temporal data (monthly trends, hourly patterns)
- Built the infographic with pure CSS animations (no JS frameworks)