{"slug": "show-hn-potential-conversation-partners-in-bluesky-using-agentic-search", "title": "Show HN: Potential conversation partners in Bluesky using agentic search", "summary": "A Show HN experiment called whalesong used an agentic search setup built on a bespoke mini-harness with Gemini Argon (low) to find conversation partners among Bluesky posts, matching 79% of a 100-post sample set to meaningful counterparts. The project indexed a one-week window of filtered Bluesky posts in a Postgres database, used an agentic process to define a conversation-seeking cohort through iterated heuristic queries, then generated and refined search queries to retrieve matches for each seed post. The best configuration consumed an average of 42K input tokens per match search, 56.8% of them cached, with negligible output tokens, leading the author to conclude that technology is not the blocker for a context-based networking experience.", "body_md": "What if instead of talking to AI, we could find human counterparts currently sharing the same interest to talk to? Some whales kind of do that by singing across the ocean.\n\nThis inspired a thought process for a conceptual social network - whalesong - where an idea, thought or a conversationally valuable statement could be matched on.\n\nIn contrast with the wide-broadcast social network visibility (restricted on groups or connections), I’d love to explore the idea of context-based networking.\n\nThis experiment is showcasing the technical feasibility and potential value in whalesong.\n\nThe goal: uncover conversational counterparts among Bluesky posts using an agentic search algorithm.\n\nThe setup: a one-week window of bluesky posts with some filtering (e.g. mostly text, no promotions or reposts/comments), indexed in a postgres database. A curated seed set - an agentic process defines a conversation-seeking cohort by creating some heuristic queries, iterating and validating it on retrieved data and resulting in a non-exhaustive set of high-quality data points. Counterpart retrieval - for a seed conversation-seeking post, an agentic process generates search queries, runs them against the database and retrieves potential matches. It reviews retrievals and iterates on improving or expanding the search.\n\nOut of several configurations, the best performing agentic setup (bespoke mini-harness with Gemini Argon - low) managed to find meaningful matches to 79% of a sample set of 100 seeds. This setup consumed on average 42K input tokens on each match search, of which 56.8% were cached, output token count was negligible.\n\nSo this experiment made it clearer to me that technology is not the blocker for a “whalesong” experience. This leads me to further investigate the experience itself. Would love to hear from you folks about experiences in getting in touch with a complete stranger on the web because both of you asked the same question or had a similar problem to solve. What made you connect? Where did it happen? Did finding similarity make connecting easy?\n\nComments URL: [https://news.ycombinator.com/item?id=50276379](https://news.ycombinator.com/item?id=50276379)\n\nPoints: 2\n\n# Comments: 0", "url": "https://wpnews.pro/news/show-hn-potential-conversation-partners-in-bluesky-using-agentic-search", "canonical_source": "https://sketch.wsng.app/", "published_at": "2026-10-28 16:32:12+00:00", "updated_at": "2026-10-28 16:42:07.270327+00:00", "lang": "en", "topics": ["ai-agents", "ai-search", "artificial-intelligence"], "entities": ["Bluesky", "whalesong", "Gemini Argon", "Postgres", "Hacker News"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/show-hn-potential-conversation-partners-in-bluesky-using-agentic-search", "markdown": "https://wpnews.pro/news/show-hn-potential-conversation-partners-in-bluesky-using-agentic-search.md", "text": "https://wpnews.pro/news/show-hn-potential-conversation-partners-in-bluesky-using-agentic-search.txt", "jsonld": "https://wpnews.pro/news/show-hn-potential-conversation-partners-in-bluesky-using-agentic-search.jsonld"}}