Hacker News: antonyragleap

Show HN: RagLeap v0.4.0 – 46 AI roles, 8 connectors, 9 vector DBs

antonyragleap — Sat, 12 Sep 2026 07:33:30 +0000

Article URL: https://github.com/antonyrag/ragleap-core

Comments URL: https://news.ycombinator.com/item?id=49669907

Points: 1

# Comments: 0

New comment by antonyragleap in "Show HN: RagLeap Core – 46 AI Employees, open-source LangChain alt"

antonyragleap — Sat, 05 Sep 2026 12:25:33 +0000

Fair! Lots of AI wrapper slop out there.

Difference here: not a wrapper - it's 4 PyPI libs (26k downloads, 245+ tests) that run fully self-hosted. ragleap-rag supports 6 vector DBs + hybrid + reranking, ragleap-graph does Neo4j. MIT, BYOK, no API call to us.

Happy to be called out - what would make this NOT slop in your view? Genuine feedback welcome.

New comment by antonyragleap in "Show HN: RagLeap Core – 46 AI Employees, open-source LangChain alt"

antonyragleap — Sat, 05 Sep 2026 06:34:16 +0000

Thanks for checking! Built this as open alt to LangChain/CrewAI - 46 role-based AI Employees that run from your own docs on your own server. Self-hosted, BYOK, MIT.

Stack is 4 PyPI packages (26k downloads) - ragleap-rag with 6 vector DBs + hybrid search, ragleap-graph with Neo4j.

Currently at 0 stars, day 0 - Would love feedback from anyone who built RAG in production - what broke for you?

Show HN: RagLeap Core – 46 AI Employees, open-source LangChain alt

antonyragleap — Sat, 05 Sep 2026 06:24:29 +0000

Article URL: https://github.com/antonyrag/ragleap-core

Comments URL: https://news.ycombinator.com/item?id=49573630

Points: 2

# Comments: 3

New comment by antonyragleap in "Rhombus 1.1 is now available"

antonyragleap — Mon, 17 Aug 2026 06:12:01 +0000

The practicality of Racket is interesting; sometimes a language wins simply because its ecosystem makes certain tasks much easier.

New comment by antonyragleap in "How I use LLMs to learn complex topics"

antonyragleap — Mon, 10 Aug 2026 05:38:28 +0000

The back-and-forth questioning seems especially useful for exposing gaps in your mental model rather than just producing a summary.

New comment by antonyragleap in "DeepSeek V4 Flash 0731"

antonyragleap — Sat, 08 Aug 2026 06:38:15 +0000

The interesting part for me is whether DeepSeek V4's reasoning gains hold up on long-running tasks rather than short benchmarks.

Show HN: Ragleap-RAG – RAG engine that documents what it doesn't do yet

antonyragleap — Tue, 04 Aug 2026 07:13:19 +0000

Article URL: https://github.com/antonyrag/ragleap-core

Comments URL: https://news.ycombinator.com/item?id=49165268

Points: 1

# Comments: 0

New comment by antonyragleap in "Agent Skill to Force Docs in ASD-STE100 Simplified Technical English"

antonyragleap — Fri, 31 Jul 2026 06:16:48 +0000

I wonder whether the real value is the shared, versioned workflow rather than the prompt itself.

New comment by antonyragleap in "LearnVector – Andrew Ng's AI company building one‑to‑one learning experiences"

antonyragleap — Wed, 29 Jul 2026 07:30:36 +0000

The hardest problem seems less like content generation and more like accurately modeling what a learner actually understands.

New comment by antonyragleap in "PGSimCity - How PostgreSQL Works"

antonyragleap — Mon, 27 Jul 2026 05:20:25 +0000

Visualizing MVCC and transaction visibility would make this even more valuable as a learning tool.

New comment by antonyragleap in "Are AI labs pelicanmaxxing?"

antonyragleap — Thu, 23 Jul 2026 05:51:34 +0000

The interesting question is whether this generalizes beyond pelicans to benchmarks the model hasn't seen before.

New comment by antonyragleap in "Your Eye Health: It's Time to Talk About It"

antonyragleap — Fri, 17 Jul 2026 04:57:56 +0000

blue light glasses did nothing for me, distance from the screen did everything.