After scathing accusations of skimping on due diligence, as well as other feedback to my article on trying to use an ‘AI coding assistant’ for the first time, the only rational, academic response is to lick one’s wounds following a particularly bruising peer review and try to address the raised issues. Reality after all does not care about one’s feelings, and there may be more to this AI assistant technology that can be coaxed out with a more in-depth look.

To this end I’ll do my best to try and work through each raised point, criticism and accusation, to see what I – and perhaps others – can learn of this endeavor. Said points include the use of the wrong frontend – i.e. Copilot – and the wrong model – being Claude Haiku 4.5 – as well as the egregious flaw on my end of ‘prompting wrong’.

For the sake of due diligence the best frontend and models will be investigated for particular tasks, with finally the verbal minefield of ‘prompt engineering’ examined for industry-standard approaches.

Junior Developer

The exact way to refer to an LLM coding assistant is still in flux, with some comparing it to pair programming, while others see the assistant more as a glorified search engine that also has code-complete features as a kind of merger of a web search engine and IntelliSense in Visual Studio. This relationship and how to look at it is the cause of a lot of contention as a result.

Another perspective is that of these assistants being more like junior developers. After all, they can apparently do all the basic boilerplate stuff, write unit tests and perform a range of other basic tasks that are beneath more senior developers. The corollary here is then of course why companies would even want to hire another junior developer if the LLM can fill these jobs. Unsurprisingly, it is already being reported that this happening.

The million dollar question that remains is that if all of this is true whether a junior developer still has value. The answer appears to be ‘yes’, even if you ask Microsoft. The argument would appear to boil down to that these assistants supposedly automate away a lot of the tedium that used to get pushed onto junior developers, leaving them free to develop more advanced skills, naturally supported by the same coding assistants.

Fancier Automation

This gets us to the question of whether these assistants are really much better than the automation tools that have existed in IDEs for many decades now with arguable improvements over time. They certainly do seem to be more capable, but they’ll still never exceed their programming, and require a lot of finagling to make them do the right thing.

Returning to junior developers for a moment: bad apples aside, they will let you know if they didn’t understand something correctly, ask for clarification, admit that they don’t know something and offer to look something up in the documentation if they do not know. None of these are things that these glorified chatbots are capable of, which makes a comparison with IDE automation tools rather fair, especially since junior developers tend to get fired if they screw up as badly as the LLM tools seem to regularly do.

While it’s true that these newfangled coding assistants do have a context window in which they “remember” previous details, you’re still dealing with the limitations of the underlying model no matter how good your prompt engineering skills are. They will also regularly confabulate and you have to accept that they generate code and documentation that is just as likely to be correct as completely wrong, even if many users of these tools seem to believe that they are actually more performant.
Self-reported and observed AI coding assistant performance. (Credit: Joel Becker et al., METR, 2025)
Ergo you’ll be writing test cases for the test cases and generated code, while also pulling code review duty, as there is no possibility of ever establishing a level of trust. Especially not after it deletes your entire hard drive or the production database for the second time that week.

If that sounds like the kind of junior developer or automation tool you’d love to be paired up with, then you’re quite the adventurous spirit. Meanwhile I have had enough fun with even code completion tools like the aforementioned IntelliSense or its equivalents in the various other IDEs that I have used over the years to never use them again. It’s bad enough when a code completion tool gets it wrong, it’s worse when the human in the loop fails to catch the glaring mistake.

Model Frontends

Although we generally refer to ChatGPT, Claude or Copilot as an LLM, this is technically incorrect, as these are merely the chatbot frontends that are written to provide a natural language interface experience. The choice here is naturally quite dizzying, as you have a range of major players including the aforementioned, each of which offers a web interface as well as integration with various IDEs and use on the CLI for easy automation.

Hence the claim that one should never use the web frontend for coding, as it needs access to your code and local environment, which makes sense if you want more of the pair programming experience. Since my ‘IDE’ of choice are Notepad++ and Vim, my options here are of course rather limited. There is a third-party OpenAI integration plug for NP++ called NPPOpenAI, but that would seem to be it.

The cool kids are of course all using Visual Studio Code with direct integration of all the frontends, but that option seems about as appealing as ripping half the RAM out of my PC and smashing my fingers with a hammer. Even as a former avid Visual Studio Pro user I feel insulted on a fundamental level at the mere thought.

Maybe that one of the CLI tools like Copilot CLI are a better match for me here as suggested, but this would appear to be more of a way to automate various GitHub tasks. Despite searching around I could not find an objective comparison of the different frontends, just many strong opinions and various pricing plans for model access, so for all intents and purposes they’re being treated as the same.

The Model Catwalk

It was further suggested that I take a look at LiveBench.ai for a comparison of how models perform on various tasks. This does indeed appear to be a valuable resource, if only for providing what appears to be a fairly objective way to compare these individual models against each other.

When sorting by the heading Coding Average, it puts OpenAI’s GPT-5.2 Codex at the top, with Claude 4.7 Opus Thinking High Effort close behind at both a hair over 83%. The Haiku 4.5 model that I was using comes in at a mere 72.17%, which is still much better than the sub-60 percent models near the bottom. Of the free models Haiku 4.5 at least would seem to be not too terrible, with Anthropic marketing it in October of 2025 as equivalent to Sonnet 4 when it comes to coding performance:
Haiku 4.5 model comparison at launch. (Credit: Anthropic)
Consequently, it would be expected to perform at least decently at given tasks, but we can take a look at what other models are available via GitHub Copilot, for instance.

If you’re not into shelling out any clams for a purported improved experience with at least the Pro+ – not Pro – subscription, you get access to quite a few models to pick from that are apparently not ‘premium’. Of note here is that new sign-ups are currently ‘paused’ as usage-based billing is being introduced.

Of these available free models the following would have theoretically performed better according to the aforementioned benchmarks:

• GPT-5.4 Mini, at 74.70%.
• GPT-5 Mini, at 76.07%.

Following the ‘fast and cost-efficient’ category things get a bit dicey to compare, due to Anthropic’s awesome naming scheme and apparently an additional mode you can use these models in, which may or may not apply here:

• Claude Sonnet 4.6 (Thinking High Effort), 79.9%.
• Claude Sonnet 4.5 (Thinking), 80.36%.

These are apparently ‘more versatile and highly intelligent’, which doesn’t seem to bump up their total score too much compared to the mini models. Following this we get the ‘most powerful at complex tasks’ models:

• GPT-5.4 (Thinking High Effort), 78.18%.
• GPT 5.3 Codex (High), 78.18%.

Taken at face value, the 72.17% for the Haiku 4.5 model is indeed somewhat worse than the other two mini models, yet as this points system relies on a specific methodology it’s important to consider what this means. From the underlying coding tests we can see that they are all Python-based programming examples, which is great if you’re testing Python coding assistants, but rather useless for my purposes as I program in just about any language except Python.

Perhaps more worrying here is the statistic that even in this scenario the best model (GPT-5.2 Codex) only managed to score a rather pitiful 83.62%, so your choice would seem to be roughly between ‘atrocious’ and ‘very bad’. Within the free model selection you’re choosing between roughly 28% and 22% of the answers being incorrect, or roughly a 3/4 chance of getting what you were asking for.

Statistically, this wouldn’t seem to make much of a difference when picking either model.

Prompt Engineering

On my last foray, I was also accused of “prompting the wrong way”, which brings us to the topic of prompt engineering, where you must learn to follow specific rules in order to “correctly” use one of these coding assistants. A crucial aspect that was not obvious to me is that you absolutely must use so-called ‘environmental prompting’, where you set the equivalent of global variables.

To this you then add , such as in the absolute gem that is used by the Livebench code test for an array test:
### Instructions: You are an expert Python programmer. You will be given a question (problem specification) and will generate a correct Python program that matches the specification and passes all tests. You will NOT return anything except for the program.
### Question:
You are given an integer array nums and an integer k.
The frequency of an element x is the number of times it occurs in an array.
An array is called good if the frequency of each element in this array is less than or equal to k.
Return the length of the longest good subarray of nums.
A subarray is a contiguous non-empty sequence of elements within an array.

Example 1:

Input: nums = [1,2,3,1,2,3,1,2], k = 2
Output: 6
Explanation: The longest possible good subarray is [1,2,3,1,2,3] since the values 1, 2, and 3 occur at most twice in this subarray. Note that the subarrays [2,3,1,2,3,1] and [3,1,2,3,1,2] are also good.
It can be shown that there are no good subarrays with length more than 6.

Example 2:

Input: nums = [1,2,1,2,1,2,1,2], k = 1
Output: 2
Explanation: The longest possible good subarray is [1,2] since the values 1 and 2 occur at most once in this subarray. Note that the subarray [2,1] is also good.
It can be shown that there are no good subarrays with length more than 2.

Example 3:

Input: nums = [5,5,5,5,5,5,5], k = 4
Output: 4
Explanation: The longest possible good subarray is [5,5,5,5] since the value 5 occurs 4 times in this subarray.
It can be shown that there are no good subarrays with length more than 4.

Constraints:

1 <= nums.length <= 10^5
1 <= nums[i] <= 10^9
1 <= k <= nums.length

### Format: You will use the following starter code to write the solution to the problem and enclose your code within delimiters.
```python
class Solution:
def maxSubarrayLength(self, nums: List[int], k: int) -> int:

```

### Answer: (use the provided format with backticks)
With this kind of preamble and explicit instructions to the ‘coding assistants’, you may as well just write the code yourself. Even if further brevity is usually ‘good enough’, the need to spend all that time and effort just to get the answer that you know you were looking for. Even a junior developer wouldn’t need this much hand holding.

In this regard, the other uses that people have mentioned, such as bouncing ideas off the chatbot — “rubber ducking” — has merit, but often even talking to potted plants or going on that walk around the block can do just about as much with getting one’s thoughts in order, along with random web searches.

Whether to use “micro prompts” or larger tasks, whether to use the chatbot as a search engine or not, and whether to correct answers to previous prompts are all details that seem to be highly divisive among users of these tools, as is the topic of vibe coding, which some seem to embrace, while others dismiss it as an insult for their artisanal craftwork.

Local Models

There are many more things to cover here, such the use of local models vs these hosted ones, with all the gotchas of subscriptions, private data harvesting and the like that this entails, but that will have to wait for another article. It’s also interesting how much the subscription and usage fees (and limitations) are currently going up across the various services, making the idea of local models seem more attractive, if they are even worth it with such limited inference capacity available.

Suffice it to say that I have learned some things along the way of writing this article, while not changing my overall premise and conclusion of the previous article. Although I could have certainly picked a theoretically better model, this is hard to to substantiate without pitting the models against each other in STM32 CMSIS and Ada coding challenges. Based on the results in Python it’s hard to make the claim that it would have made an amazing distinction, but maybe not using a ‘mini’ model makes all the difference here?

Hopefully the better models won’t be removed from free access before I can even give this idea a shot.

hackaday.com/2026/06/08/revisi…

Se vivi nella linea temporale più stupida del multiverso, la stupidità non è una scusa, è una spiegazione...

dk.dataknightmare.eu/abbiamo-e…

dataKnightmare Archive

2026-06-08 12:42:24

DK 10x34 - Abbiamo escluso la stupidità?

Ascolta l'episodio su Spreaker.com

Il teatrino del digitale sembra ogni giorno di più la Nave dei Folli, al punto che a volte dubito perfino che le mie pur sovrumane capacità di trovare un filo conduttore possano bastare.

Per fortuna, quando rischio di essere travolto dal dubbio, mi ricordo un altro dei miei capisaldi culturali, le vignette dei Peanuts, in particolare quella in cui Lucy nei panni della psicologa, dovendo spiegare le ragioni di non ricordo cosa dice:

"abbiamo escluso la stupidità?"

Ecco, questa è una chiave di lettura in cui si allineano armoniosamente un sacco di pezzi sparsi:

• ...aziende che non hanno mai generato un profitto, con perdite per miliardi, che esistono da anni solo grazie al continuo sussidio di capitali di ventura
• ...l'intera economia americana sono sette aziende che si passano l'un l'altra un "pagherò" da cento miliardi di dollari
• ...Microsoft confessa che l'Intelligenza Artificiale le costa più che pagare le persone
• ...un'azienda si sveglia una mattina e scopre di avere speso 500 milioni su Claude in un mese
• ...i giganti del digitale sono in mano a gente a cui non affideresti l'imbiancatura del salotto.

Ecco, siamo sicuri di avere escluso la stupidità?

Perché se dobbiamo credere, e ci crediamo per esperienza, al principio di Peter secondo cui ciascuno viene promosso fino al livello della propria incompetenzza, allora la stupidità non è una scusa, è una spiegazione di quello che stiamo vedendo.

Ed è anche l'enorme consolazione di sapere che non siamo noi ad esserci bevuti il cervello, siamo solo in mano a un branco di imbecilli.

Come altro spiegare una delle manovre di marketing più fantasmagoriche della storia, quella di Anthropic?

• Anthropic il giorno 1 lascia i propri codici sorgente a disposizione del primo che passa,
• il giorno 2 tutti si accorgono che il codice di Claude è una montagna di sterco glassato, non credete a me, andate pure a leggervi le analisi
• il giorno Anthropic 3 annuncia un nuovo modello, Mythos, "troppo pericoloso per il rilascio al pubblico" accompagnato dalle solite statistiche inventate
• e il giorno 4 tutti parlano esclusivamente di Mythos come se facesse qualcosa che non era possibile anche prima, anche con altri modelli linguistici, e anche senza modelli linguistici.

Davvero, leggete qualsiasi analisi e a meno di essere sotto ketamina ne concluderete che il sorgente di Claude è pura programmazione a martellate intervallata da inviti al modello linguistico a non fare cazzate, per favore.

Se a questo aggiungete il fatto che per generare profitti Anthropic, ma anche openAI, ma anche Google, dovrà a un certo punto aumentare i prezzi della cosiddetta intelligenza artificiale di quasi venti volte, capite da soli che l'inevitabilità dell'Intelligenza Artificiale è una proposizione di marketing, e nient'altro.

Prima ancora di qualsiasi discussione su cosa veramente possano fare o non fare i modelli linguistici, facciamoci una domanda molto capitalista: sono un modello di business in grado di funzionare oppure no?

Perché messa in questi termini la domanda, la risposta è: "no".

Perché a un certo punto gli investitori si stuferanno di veder bruciare i loro soldi a sussidio dei prezzi di vendita.

Perché a un certo punto smetteranno di costruire i datacenter a debito nell'attesa della meravigliosa età dell'oro.

Perché a un certo punto la sbobba prodotta dai modelli linguistici avrà un peso tale nei dati di training che i modelli cominceranno a collassare. Tutti, e tutti assieme.

L'ho già detto: questa mania dell'intelligenza artificiale fallirà miseramente, e il solo problema è che fallirà economicamente prima di poter essere demolita culturalmente, il che significa che fra trent'anni o giù di lì, tornerà ad infestarci.

Possiamo solo fare quello che dobbiamo fare adesso, e sperare che i nostri figli siano preparati.

pausa

Nel frattempo, per non farci mancare niente, spuntano come funghi titoli come "la rivolta contro l'intelligenza artificiale", visto sull'ultimo Internazionale.

Questi media fanno tenerezza, e poi si lamentano che nessuno li legge.

Nel caso di Internazionale, l'articolo di copertina non parla di una rivolta, ma di una manifestazione di protesta, e non contro l'intelligenza artificiale, qualunque cosa sia, ma contro la costruzione di datacenter.

Detto in parole povere: gente che non vuole un datacenter vicino a casa propria per quattro motivi: inquina, consuma tutta l'acqua potabile, deprime il valore degli immobili, e fa aumentare le bollette perché le aziende di energia non scaricano i costi di espansione dell'offerta sulle aziende di AI che consumano come intere città, ma sulle utenze domestiche. Perché? Perché le utenze domestiche non possono minacciare di costruirseli da soli, i datacenter, mentre i techbro, sì.

"Viviamo nella linea temporale più stupida del multiverso" l'abbiamo già detto?

Se aveste ancora qualche dubbio, l'ultima tendenza delle agenzie di intelligence è giustificare la propria esistenza con una nuova minaccia inventata: l'estremismo anti-tech.

Visto che ormai anche mia nonna ha capito che il terrorismo transnazionale è quello che fa Israele a Gaza e in Libano, la Russia in Ucraina, e gli USA in Iran, le agenzie hanno trovato una nuova minaccia low-cost che gli permette di usare tutti i loro costosissimi giocattoli senza avere a che fare con gente capace davvero di sparargli addosso: nientepopodimeno che "l'estremismo anti-tech" ^[1][2], ossia tutti quei pericolosi figuri che prendono un pennarello e un cartello e scendono in piazza a dire che non è carino licenziare la gente e bruciare il pianeta per far arricchire Musk, Amodei e Altman.

Considerato che le squadre di polizia antisommossa sono specializzate nel randellare studenti e pensionati, non fa una piega.

Scherzi a parte, questa è una cosa su cui ragionare: le tecnologie hanno una propria politica^[3], lo sappiamo da quasi cinquant'anni, anche se puoi laurearti in ingegneria o in informatica senza che nessuno te lo spieghi.

Dopo l'11 settembre, gli Stati Uniti, completamente incapaci di capire cosa fosse successo e perché, hanno ampliato a dismisura i propri apparati di sorveglianza, e li hanno sguinzagliat contro palestre, organizzazioni sindacali, gruppi pacifisti, chiunque non applaudisse ciecamente la linea governativa che vedeva gli USA come povera vittima innocente, colpita perché incarnava i sacri valori di Mercato e Libertà, rigorosamente in quest'ordine.

Venticinque anni dopo vediamo benissimo che quella fu una involuzione fascista, di cui oggi si vedono i risultati interni ed esterni.

Il problema è che l'Europa, come sempre, è andata al seguito, e oggi anche da noi la deriva fascistoide comincia a essere percepibile:

• nella retorica contro i migranti (che se si smettesse di bombardarli non migrerebbero),
• nella russofobia d'accatto in una questione ucraina ingegnerizzata e finanziata dagli Stati Uniti,
• e nella retorica militarista di una Europa che dovrebbe investire in armamenti per "avere maggior peso internazionale", naturalmente senza dire quando, come e contro chi questi armamenti dovrebbero essere usati, e soprattutto chi nome e cognome dovrebbe andare eventualmente al fronte.

Possiamo anche vivere nella linea temporale più stupida del multiverso, ma stupido non significa innocuo. Apriamo un po' gli occhi.

pausa

Veniamo a questioni pratiche, il libro di Weizenbaum: l'editore mi ha fatto il simpatico scherzetto di scoprire dopo sette mesi che non ha più un contratto con il traduttore.

Ho già debitamente invocato numerose divinità colpevolmente dimenticate dalla storia, ma a questo punto mi sono stufato di perdere tempo. Il traduttore dell'edizione è un certo Federico Tibone, che purtroppo non conosco.

Ma si dà il caso che abbia prodotto quasi una ventina di libri con Zanichelli, che non è un cattivo nome nell'editoria.

Quindi ho già scritto a Zanichelli, ho spiegato la situazione, e ho chiesto che mi mettano in contatto con Tibone per fare, a questo punto, una ristampa del Weizenbaum con Zanichelli.

Non ho illusioni sui tempi aziendali di un gigante del genere. Ma, per una volta, non credo di essere solo. Se conoscete Tibone, o se conoscete un dirigente in Zanichelli che sappia prendere una decisione in meno di un'era geologica, metteteli in contatto con me.

Se non conoscete nessuno, spendete due minuti e scrivete a Zanichelli chiedendo quando pubblicheranno "Il potere del computer e la ragione umana" di Joseph Weizenbaum.

Fatevi sentire, rompete le palle. Perché giuro, a costo di ritradurlo io quest'estate e fare un'autoproduzione, a Pordenone il Weizenbaum per il di ELIZA ce lo porto.

A questo punto, è una questione di principio.

---

1. Feds Targeting Anti-Tech Extremism As AI Backlash Spreads↩︎, The New American
2. US Law Enforcement Warns Of Anti-Tech Extremism, Wired.com↩︎
3. Do Artifacts Have Politics?, Langdon Winner, 1980↩︎

US Law Enforcement Warns of ‘Anti-Tech Extremism’ as AI Hatred Grows

As Americans stew over the looming risk of job-stealing AI and data centers in their back yards, the feds are raising the alarm about a new category of threat, documents obtained by WIRED show.

^{Daniel Boguslaw (WIRED)}

IT'S MONDAY, AND THIS IS DIGITAL POLITICS. I'm Mark Scott, and I'll be speaking at this Harvard University/IAPP digital conference in New England at the end of the month. If you're in Boston and want to meet, let me know here.

— The European Union, Canada and the United States outlined dueling visions for sovereign AI. None of them is keeping up with industry.

— Australia laid out a game plan for social media bans aimed at children. The country's early findings demonstrate the policy doesn't address the root cause of parents' anxieties.

— Two-thirds of future US data centers will be built in rural areas.

Let's get started:

digitalpolitics.co/sovereign-a…

There’s not much more nostalgic for many than good ole’ Jimmy Neutron. This was true for [Kiara], who saw the gorgeous pupper Goddard and wanted him for herself. Of course, there was no solution other than to make an animatronic version of the robot dog.

Starting with some files ripped from a Jimmy Neutron GameCube game, Goddard was designed digitally before being printed in life size. Of course, for a true reproduction of the robot dog, the parts had to be prepped and painted in the iconic chrome and purple. A real plasma ball was used for the brain, and linear actuators were used for the legs. The head was able to be moved around similarly to professional animatronics using fishing line and servos. Put together the entire finished pup, looks incredible.

[Kiara] isn’t a stranger to the animatronic world, and Hackaday isn’t a stranger either. If you want to see some other funky animatronics to bring fiction to life make sure to check out the singing Magikarp. Or if you are wanting something a bit more serious, check out the dystopian work from Danny Huynh!

youtube.com/embed/iMNQjPxvfTk?…

hackaday.com/2026/06/08/goddar…