Hey PH, Built this for myself. The problem isn't that there's too much AI news, it's that it's scattered everywhere with no good way to get the signal without the noise. HACKOBAR_ pulls from HN, arXiv, GitHub, HuggingFace, Reddit, lab blogs, newsletters and Twitter into one feed. Scoring weighs engagement, recency and cross-platform presence, so a paper that shows up on arXiv, gets posted to HN and picked up in a newsletter ranks higher than something that just went viral on one platform. Source authority weights stop social volume from drowning out research. The LLM rates each story's significance independently so a quiet but important paper isn't buried by a loud but shallow one. Summaries are hype-free by design. No "groundbreaking", no "revolutionary". Still early. Would love to know what sources you think are missing and whether the scoring feels right. - Rahul

HACKOBAR_

One feed for AI signal. No noise.

One feed for AI signal. No noise.