Hacker News: lyjackal

New comment by lyjackal in "Gemini 3.5 Flash"

lyjackal — Tue, 19 May 2026 18:29:38 +0000

You’re quoting the batch pricing. On demand is 1.5 per input and 9 per M output. This is effectively comparable cost to Gemini 2.5 Pro in a flash tier model

New comment by lyjackal in "Claude Design"

lyjackal — Fri, 17 Apr 2026 18:56:34 +0000

I think the larger part implied is the design will be crappy, because the problem was unexplored

New comment by lyjackal in "How we give every user SQL access to a shared ClickHouse cluster"

lyjackal — Sat, 21 Mar 2026 14:41:37 +0000

I want to build a shared postgres db with hundreds of small apps (OLTP) accessing shared tables using a RLS model model against well defined tables.

What are other limitations and mitigations folks have used or encountered to support stability and security? Things like

  - Query timeouts to prevent noisy neighbors
  - connection pooling (e.g. pgbouncer) also for noisy neighbors
  - client schema compatibility (e.g. some applications running older versions, have certain assumptions about the schema that may change over time)

New comment by lyjackal in "Show HN: Local Document Parsing for Agents"

lyjackal — Thu, 19 Mar 2026 18:55:55 +0000

github repository here https://github.com/run-llama/liteparse

New comment by lyjackal in "SQLite's documentation about its durability properties is unclear"

lyjackal — Fri, 29 Aug 2025 18:33:43 +0000

it's Friday, and I'm avoiding doing work

New comment by lyjackal in "Workflows 1.0: A Lightweight Framework for Agentic systems"

lyjackal — Mon, 30 Jun 2025 20:12:00 +0000

I notice that the python versions and typescript versions are pretty different. Python is sort of class based, with python magic decorators

    class MyWorkflow(Workflow):
        @step
        async def start(self, ctx: Context, ev: StartEvent) -> MyEvent:
            num_runs = await ctx.get("num_runs", default=0)

whereas TS is sort of builder/function based

    import { createWorkflow } from "@llamaindex/workflow-core";
    
    const convertEvent = workflowEvent();
    
    const workflow = createWorkflow();
    
    workflow.handle([startEvent], (start) => {
      return convertEvent.with(Number.parseInt(start.data, 10));

Is there reason for this? });

New comment by lyjackal in "MCP Specification – version 2025-06-18 changes"

lyjackal — Thu, 19 Jun 2025 01:36:36 +0000

Agree, thanks for the link. I was wondering what actually changed. The resource links and elicitation look like useful functionality.

New comment by lyjackal in "Reverse Engineering Cursor's LLM Client"

lyjackal — Sat, 07 Jun 2025 16:01:11 +0000

I've been curious to see the process for selecting relevant context from a long conversation. has anyone reverse engineered what that looks like? how is the conversion history pruned, and how is the latest state of a file represented?

New comment by lyjackal in "Python Tooling at Scale: LlamaIndex’s Monorepo Overhaul"

lyjackal — Wed, 21 May 2025 18:22:49 +0000

I recently did something similar. Using uv workspaces, I used the uv CLI's dependency graph to analyze the dependency tree then conditionally trigger CI workflows for affected projects. I wish there was a better way to access the uv dependency worktree other than parsing the `tree` like output

New comment by lyjackal in "Show HN: Juvio – UV Kernel for Jupyter"

lyjackal — Tue, 20 May 2025 18:26:04 +0000

This is cool and something that I’ve wanted, but I don’t see hot listings requirements inline foregoes the need for a lock file to maintain reproducibility. What about version ranges? Versions of transitive dependencies?

New comment by lyjackal in "Scraperr – A Self Hosted Webscraper"

lyjackal — Mon, 12 May 2025 02:36:54 +0000

Mobile 4g USB sticks you can usually rotate your IP address by reconnecting. I tried on a pi, it was inconsistent. This was just with some random test mobile plan from rando carrier renting off Verizon I think

New comment by lyjackal in "Ask HN: What are you working on? (April 2025)"

lyjackal — Mon, 28 Apr 2025 11:03:33 +0000

An idea for more complete coverage: have 2 of them, and invert their intervals, such that one and only one is always on

New comment by lyjackal in "Should We Respect LLMs? A Study on Influence of Prompt Politeness on Performance"

lyjackal — Tue, 22 Apr 2025 13:27:02 +0000

I’ve wondered whether they use thanks as a signal to a conversation well done, for the purpose of future reinforcement learning

New comment by lyjackal in "Adobe deletes Bluesky posts after backlash"

lyjackal — Fri, 11 Apr 2025 22:56:16 +0000

It's more the content creators who bear the brunt of toxic rage. Who you follow doesn't solve that problem

New comment by lyjackal in "Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison"

lyjackal — Mon, 31 Mar 2025 12:46:32 +0000

https://aistudio.google.com

New comment by lyjackal in "Posthog/.cursorrules"

lyjackal — Sun, 09 Mar 2025 12:02:21 +0000

Yes, I run into it, but it’s intermittent. Cursor makes some internal decisions to limit its context budget, so my speculation is that it’s related to that, like the actual updated code is just not in the prompt sometimes

New comment by lyjackal in "Ingesting PDFs and why Gemini 2.0 changes everything"

lyjackal — Thu, 06 Feb 2025 02:10:51 +0000

If the end goal is just rag or search over the pdfs, seems like ColPali based embedding search would be a good alternative here. Don’t process the PDFs, instead just search their image embedding directly. From what I understand, you also get a sort of attention as to what part of the image is being activated by the search.

New comment by lyjackal in "Diagram as Code"

lyjackal — Tue, 05 Nov 2024 20:20:54 +0000

drawio and its vscode plugin supports the same

New comment by lyjackal in "Play 3.0 mini – A lightweight, reliable, cost-efficient Multilingual TTS model"

lyjackal — Mon, 14 Oct 2024 21:54:29 +0000

Is there any way to use the TTS on its own? I maintain an obsidian TTS plug-in, and am starting to add new TTS providers (its just been OpenAI thus far). From the documentation at https://docs.play.ai/documentation/get-started/introduction, it looks like their API seems to couple it to an LLM for building conversational agents. Seems like it might be nice to use standalone as just TTS.

New comment by lyjackal in "Parler TTS v1"

lyjackal — Fri, 09 Aug 2024 13:38:31 +0000

Apologies for the x link. I couldn’t find a more official announcement than this. Yesterday hugging face released new checkpoints with much higher quality output, trained for longer, and with speaker consistency.

Demo here: https://huggingface.co/spaces/parler-tts/parler_tts

GitHub: https://github.com/huggingface/parler-tts