zarldev
Re:How I Fight AI Brain Fry
on Sunday November 08, 2026 @08:33PM (#50419862)
Brittle by building ever more increasing numbers of abstractions rather that take a coherent view of some things. So over time you have these "seams" that the code base ends up looking like a punk motor cycle jacket.
Re:How I Fight AI Brain Fry
on Sunday November 08, 2026 @08:28PM (#50419813)
Event Catalogue is something I have been using for this https://www.eventcatalog.dev/ only thing is no longer MIT license.
Re:What Is a Harness?
on Monday August 24, 2026 @07:47AM (#49418379)
A harness is a tool that allows the LLM to focus on the primary objective that the user wants, coding, research, reports, finance anything really but the harness structures the agent interactions with the supporting tools to achieve the narrow scope. Driving the LLM with correct feedback from the domain knowledge provided by the tools. TBH when I was writing my harness this is why i have the Guardrails concept around Any tool, pre and post tool call, pre can make sure the data is formatted correctly lowering incorrect tool calls and post tool calls for validaition of results of output. Harness: https://github.com/zarldev/zarlmono/tree/main/zarlcode Guardrail Design: https://zarldev.github.io/zarlmono/guardrails/ All Docs: https://zarldev.github.io/zarlmono/
Re:Building an Advanced Agentic Harness
on Wednesday August 05, 2026 @04:02PM (#49188204)
OpenAI and Anthropic both have blog posts about programmatic tool use. Tbh I settled in STARLARK as the intermediary language it's a python dialect so most llms can use it with little help.https://github.com/zarldev/zarlmono/blob/main/zkit%2Fagent%2... it's a wrapper around the calls of the other tool set.
Re:SearXNG: A free internet metasearch engine
on Saturday July 04, 2026 @07:31AM (#48784614)
https://www.zarl.dev/posts/hal-by-any-other-name Here is my write up on my local model setup also have https://zarldev.github.io/zarlmono/ as my local 1st coding agent
Re:How OpenAI delivers low-latency voice AI at scale
on Monday May 04, 2026 @05:54PM (#48015514)
Yeah Gemma4 was and is great fun to do this with - I too am building pretty much the same as yourself in Go.
https://github.com/zarldev/zarl & https://www.zarl.dev/posts/hal-by-any-other-name
Re:Structured Errors in Go
on Monday June 02, 2025 @04:38AM (#44156811)
I stand corrected. Still you cannot implement an Is on a sentinel error and is only applicable to concrete error types. And if I'm using concrete types I'm going to us errors.As
Re:Structured Errors in Go
on Monday June 02, 2025 @04:33AM (#44156779)
In standard go you would not implement Is but would use the errors.Is call. How would this work if it is a sentinel error as you would use with errors.Is??
Re:Goenums: Type Safe Enum Generator for Go
on Sunday April 27, 2025 @06:39AM (#43810893)
>That's it mostly...
Whole kubernetes, docker, traefik and many other cloud native tools are written in go.
Embedding React in Go: Another over-engineered blog
For Fakes Sake – Testing in Go
Re:Go Enums Still Suck
on Thursday March 28, 2024 @08:14AM (#39850258)
Yeah extending the struct tags to iota definitions would have made things much better.
Re:Go Enums Still Suck
on Thursday March 28, 2024 @08:01AM (#39850136)
the values are specified in the definition of the iota type.
``` type planet int // Gravity[float64],RadiusKm[float64],MassKg[float64],OrbitKm[float64],OrbitDays[float64],SurfacePressureBars[float64],Moons[int],Rings[bool] ```
Re:Go Enums Still Suck
on Thursday March 28, 2024 @07:59AM (#39850125)
This should be better now.
Re:Go Enums Suck
on Friday March 01, 2024 @03:50PM (#39566644)
But all go generate does is run a binary like stringer? This can be used in go generate.
Re:Go Enums Suck
on Friday March 01, 2024 @03:34PM (#39566483)
DSL? The go way is to use go generate I would say and it can be used with go:generate, I would like to use the AST lib to parse go files to remove the need for any json and to be more like the cmd stringer tool.
Re:Go Enums Suck
on Friday March 01, 2024 @03:29PM (#39566425)
Hey did you read to the bottom of the article. It shows that that was just a temporary solution, the real invalid code is
func (t operation) IsValid() bool { _, ok := strOperationMap[s] return ok }
Tbh it has been updated to us an array instead for string indexing.