Hacker News: tontinton

New comment by tontinton in "Previewing GPT‑5.6 Sol: a next-generation model"

tontinton — Fri, 26 Jun 2026 18:19:19 +0000

Yep this is a glimpse into the future of 500+ t/s, which is in my opinion the next big thing that validates Jevon's paradox (the models are already smart enough)

New comment by tontinton in "The Token Compression Illusion: Why I'm Skeptical of RTK"

tontinton — Fri, 19 Jun 2026 07:21:49 +0000

I meant harnesses / TUIs, sorry to confuse.

Using fd/rg sounds interesting, honestly it would require little tweaks to the bash tool lua plugin, either add to the description to prefer these binaries instead or something like that.

In general though I much prefer "advising" and encouraging the LLM to use the native tools like grep l/glob, they are implemented to be super fast, and you will get better parser output.

New comment by tontinton in "The Token Compression Illusion: Why I'm Skeptical of RTK"

tontinton — Fri, 19 Jun 2026 07:16:13 +0000

Create an issue! :)

New comment by tontinton in "The Token Compression Illusion: Why I'm Skeptical of RTK"

tontinton — Thu, 18 Jun 2026 22:23:12 +0000

It's a TUI you're right, but it's also a harness.

As much as I hate to admit, T the tools you provide, the descriptions, and prompts, all amount to pretty big changes in experience, even using the same models.

New comment by tontinton in "The Token Compression Illusion: Why I'm Skeptical of RTK"

tontinton — Thu, 18 Jun 2026 21:55:59 +0000

The biggest ones are: using tree-sitter to index code files as a tool, code_execution tool running a workflow of tools inside a python interpreter (monty), and not being a harness developed by the company profiting from selling you the shovels (and introducing "dynamic workflows" aka spawning 50 agents).

New comment by tontinton in "The Token Compression Illusion: Why I'm Skeptical of RTK"

tontinton — Thu, 18 Jun 2026 21:53:29 +0000

Enjoy, I can't go back to other agents now, too spoiled by the speed

New comment by tontinton in "The Token Compression Illusion: Why I'm Skeptical of RTK"

tontinton — Thu, 18 Jun 2026 21:01:41 +0000

Totally wrong, you underestimate the frontier's incompetence in anything other than building LLM models (ehm ehm flickering TUI for a year "written like a game engine").

I ran a bunch of benchmarks and there are proven ways to reduce tokens while achieving the same results (finding the same CVEs / finding the same bugs in CRs, etc...).

See https://maki.sh, it's my own little proof.

New comment by tontinton in "DeepSeek reasonix, DeepSeek native coding agent with high caching and low cost"

tontinton — Sun, 24 May 2026 18:10:25 +0000

Yep exactly my thoughts, went and looked at the code for the deepseek provider in my coding agent. and basically all of what the author wrote there is implemented... http://github.com/tontinton/maki for the curios

New comment by tontinton in "Show HN: Superlog (YC P26) – Observability that installs itself and fixes bugs"

tontinton — Tue, 19 May 2026 16:18:14 +0000

What's your moat?

New comment by tontinton in "Zerostack – A Unix-inspired coding agent written in pure Rust"

tontinton — Sun, 17 May 2026 17:55:36 +0000

Yo that's really similar to my very own https://github.com/tontinton/maki only I'm MIT and you're GPL, cool

New comment by tontinton in "If AI Writes Your Code, Why Use Python?"

tontinton — Mon, 11 May 2026 21:08:16 +0000

Also easier to ship a binary like a cli

New comment by tontinton in "An update on recent Claude Code quality reports"

tontinton — Thu, 23 Apr 2026 19:32:11 +0000

or you can use a non vibe designed efficient Rust TUI coding agent made by yours truly, all my coworkers use it too :) called https://maki.sh!

lua plugins WIP

New comment by tontinton in "Maki – the efficient coder (AI agent)"

tontinton — Sat, 11 Apr 2026 11:54:35 +0000

Thanks!

Currently the subagent chat windows don't allow to inject user messages like the main window, I want to change that soon though.

Regarding tiered models, it currently caps the model use to the current tier you're on, so no it can't upgrade from haiku to opus suddenly. The reasoning for that is that if you selected haiku, you probably don't want to pay for opus by accident.

New comment by tontinton in "I've sold out"

tontinton — Wed, 08 Apr 2026 19:20:51 +0000

Yes, you can use the dynamic provider to work with any local model you want (see scripts/providers)

New comment by tontinton in "I've sold out"

tontinton — Wed, 08 Apr 2026 13:40:10 +0000

Is this a good time to tell you guys that there are many more alternatives, like my very own https://maki.sh?

Shameless plug, yes, but it's free, so I think it's fair :)

New comment by tontinton in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"

tontinton — Tue, 07 Apr 2026 13:00:22 +0000

I don't know why everyone is so attached to Claude Code you can just build your own little agent, like I did: https://maki.sh/

It will 100% be better than the 500k lines of code junk that is CC.

New comment by tontinton in "Maki the efficient AI coder – Rust TUI (saves 40% tokens and low RAM)"

tontinton — Thu, 02 Apr 2026 12:50:18 +0000

Something I've been working on out of frustrations with the existing AI coder CLIs

Maki the efficient AI coder – Rust TUI (saves 40% tokens and low RAM)

tontinton — Thu, 02 Apr 2026 12:50:18 +0000

Article URL: https://maki.sh

Comments URL: https://news.ycombinator.com/item?id=47613768

Points: 5

# Comments: 1

New comment by tontinton in "1M context is now generally available for Opus 4.6 and Sonnet 4.6"

tontinton — Sat, 14 Mar 2026 18:48:50 +0000

Yeah we all converge to the same workflow, in my ai coding agent I'm working on now, I've added an "index" tool that uses tree-sitter to compress and show the AI a skeleton of a code file.

Here's the implementation for the interested: https://github.com/tontinton/maki/blob/main/maki-code-index%...

New comment by tontinton in "Show HN: Context Gateway – Compress agent context before it hits the LLM"

tontinton — Fri, 13 Mar 2026 19:33:18 +0000

Is it similar to rtk? Where the output of tool calls is compressed? Or does it actively compress your history once in a while?

If it's the latter, then users will pay for the entire history of tokens since the change uncached: https://platform.claude.com/docs/en/build-with-claude/prompt...

How is this better?