hmm.. what about supply chain security?
no thanks. i use AI.
DeepSeek Harness developer preview
201–210 of 346 posts
Re: DeepSeek Harness developer preview
#202Earlier quoted context omitted.
I promise this isn't meant to be snarky, but is that not just...logs?
It's only logging if an obsolete human does it. But the future is here and thus it's called "Agentic causality's reified temporal traceability."
"Temporally Reified Agentic Causality Traceability Report" - TRACTR.
Re: DeepSeek Harness developer preview
#203Earlier quoted context omitted.
npm as a distribution tool works well and typescript has types. Any reason why it should not be written in nodejs?
I always thought nodejs was a weird choice for CLI tools. For web stuff, sure. But for CLI, it never made sense to me. Especially when Python and Go exist.
> But for CLI, it never made sense to me. Especially when Python and Go exist.
But why? Not saying node is better, just want to know where you are coming from for my own knowledge.
Bc I would have picked typescript + node too. It has types (where python just has type hints) and a lot of developers know it already (where go is more niche).
Re: DeepSeek Harness developer preview
#204What's buried under the lede: this harness is using Cordis v4 (the paper that dropped today). Cordis has already been used for four years in a different project called Koishi that uses v3. Cordis itself is a way of hot loading and unloading plugins without restarting a running process. The cool part is that when it unloads it can revert any state and side effects it created, cleaning up its connections, memory alloca…
Sounds very cool! What do you mean by “revert side effects it created”?
Anythign that needs to be cleaned up or undone goes in ctx.effect. It returns the "inverse" (the cleanup function) when the plugin loads. Cordis then stores it and runs it when the plugin unloads. Take a look at 5.1.1 in the paper.
Re: DeepSeek Harness developer preview
#205There is such a clear lack of innovation drive in this field. Every lab just copies what the other does. One of the most baffling things to me is how the once-upon-a-time good developer instinct to make everything reusable, testable, and deterministic is just getting lost into a sea of markdown begging a language model to please act a certain way. For example this repository has a "skill" definition that consists in…
This is exactly how almost every field evolves. We like to think we are so innovative. We aren’t. When someone finally does something slightly different that’s actually useful everyone comments how that was so obvious all along and they could easily have done that too . Hindsight and all. And for every new tech , people will always try to use it for everything, no matter how unsuitable, until the hype goes down and w…
Using git hooks vs an LLM to run pre-commit checks does not require hindsight.
Re: DeepSeek Harness developer preview
#206"Every run is traceable Everything the model sees is recorded in an append-only session log: system prompts, reasoning, tool calls and results, subagent scheduling, and every context injection. In the Trajectory view, you can inspect these records by source. Resume, fork, search, and replay all operate on the same event stream." That's a killer feature, IMHO, and one that US models won't allow you to do, as their tra…
For the US models, you can look at the rewritten CoTs or just ask them what they think is happening.
Re: DeepSeek Harness developer preview
#207Earlier quoted context omitted.
He explicitly asked for feedback
Exactly. “Tell me more about X” is an open ended question, not feedback.
You're absolutely right — "tell me more about X" is phrased as an imperative, not a question. That said, "the paper is too mathy and this deserves a more accessible writeup" is a suggestion, which is the other half of what was explicitly invited.
Re: DeepSeek Harness developer preview
#208Earlier quoted context omitted.
New coding harness that seems to have some novel concepts and one of the pretty cool things on their landing page for it here: https://deepseek.com/harness/en/ is the Every Run is Traceable view: "Everything the model sees is recorded in an append-only session log: system prompts, reasoning, tool calls and results, subagent scheduling, and every context injection. In the Trajectory view, you can inspect these records…
I promise this isn't meant to be snarky, but is that not just...logs?
It's logs of activity the US models hide from you in fear of them being used by competitors.
Re: DeepSeek Harness developer preview
#209"Every run is traceable Everything the model sees is recorded in an append-only session log: system prompts, reasoning, tool calls and results, subagent scheduling, and every context injection. In the Trajectory view, you can inspect these records by source. Resume, fork, search, and replay all operate on the same event stream." That's a killer feature, IMHO, and one that US models won't allow you to do, as their tra…
For the US models, you can look at the rewritten CoTs or just ask them what they think is happening.
Asking them what they think is happening is actually not reliable, though? They don't always know how they came to a conclusion after the fact. It is probable that it's roughly similar to the path they took to get there, since it's the same weights, but it's not certain. And, if you make it standard practice to always collect that data (e.g. if you have an automated tool to ask the model to explain itself after every action to log it), it seems like you might find yourself being blocked for violating terms of service. It looks like "distilling".
In short, there are workarounds, but they're not guaranteed to work forever and they're likely to bump into terms of service.
Re: DeepSeek Harness developer preview
#210"Every run is traceable Everything the model sees is recorded in an append-only session log: system prompts, reasoning, tool calls and results, subagent scheduling, and every context injection. In the Trajectory view, you can inspect these records by source. Resume, fork, search, and replay all operate on the same event stream." That's a killer feature, IMHO, and one that US models won't allow you to do, as their tra…
That is precisely how the dreamcoder [0] ai agent I built also works.
It has an event sourced architecture in SQLite and it resolves queries using recursive CTEs (and sneaky projections to speed things up) to deliver exactly that. Identical, stable message chains to AI and complete introspection.
Bonus points include a constraint-satisfaction solver for the tiling window manager so windows never shrink too small to read. And many other keyboard-friendly features.
[0] https://www.dreamcoder.ai/ [1] https://www.dreamcoder.ai/assets/graph.webp
