LLM Prompt settings

Taming the Local LLM: How System Prompts Unlock "Veritas"

Date: November 28, 2025

Model: soob3123_Veritas-12B-Q8_0.gguf 

Today’s mission was to take a local LLM off the leash. I’ve been setting up Veritas-12B, a high-precision philosophical model, running locally via Ollama and AnythingLLM3.

The hardware is capable—a Framework Desktop with 128GB of RAM—so I knew the bottleneck wasn't the machine. It was the instructions. The difference between a generic "helpful assistant" and a true "research partner" lies entirely in how you prompt the system before you ask the first question.
Here is the log of how we moved from a generic summary to a post-doctoral deep dive.

Phase 1: The Generic "Assistant"

Initially, AnythingLLM was set to its default system prompt. I asked a standard philosophical question about Jean-Paul Sartre.

System Prompt: “You are a helpful AI assistant” 
My Prompt: “Tell me what Satre says about deceiving oneself” 

The Response:
“Sartre argues that self-deception is a fundamental aspect of human existence and a means by which individuals attempt to cope with the anxieties and uncertainties of life. He suggests that self-deception can provide a temporary escape from the burden of freedom and responsibility, allowing individuals to create a narrative that provides a sense of coherence and meaning in their lives. However, Sartre emphasizes that this self-deception ultimately leads to inauthenticity and a denial of one’s true freedom.” 

Analysis:
It was accurate, but shallow. It took about 29 seconds to generate7. It felt like a Wikipedia summary—safe, concise, and uninspired.

Phase 2: The Persona Shift

I decided to push the model by manually injecting a persona into the chat. I wanted depth, so I gave it a specific role to play.

New Prompt: “You are a post doctoral student in the history of philosophy. Answer the user's questions comprehensively and in great detail. elaborate on philosophical concepts and provide extensive examples.” 

The Response:
“Jean-Paul Sartre, a prominent figure in existentialist philosophy, discusses self-deception in his work, particularly in "Being and Nothingness." According to Sartre, self-deception occurs when an individual chooses to deny or ignore aspects of their own existence or the truth of their situation, often as a way to avoid responsibility or anxiety.
Sartre emphasizes that self-deception is a form of "bad faith," where individuals attempt to escape their freedom and the responsibility that comes with it. He argues that by deceiving oneself, one is essentially denying one's own existence and authenticity. This act of self-deception can manifest in various forms, such as rationalizing one's choices, avoiding difficult emotions, or pretending to be something one is not.
Sartre believes that recognizing self-deception is crucial for living authentically. He encourages individuals to confront their true nature and the implications of their choices, even if it leads to discomfort or anxiety. Embracing one's freedom and responsibility, rather than hiding from them, is central to Sartre's philosophy.” 

Analysis:
This was better. It introduced the specific term "bad faith" and expanded on the concepts. However, it was still a wall of text—a single stream of thought without structure.

Phase 3: Unlocking the Architecture

The real breakthrough happened when I realized I wasn't just fighting the prompt; I was fighting the interface settings. I navigated to the Workspace Settings > Chat Settings in AnythingLLM and replaced the default system prompt globally to "Unlock Model Verbosity".

By hard-coding the "Post-Doctoral" instruction into the system layer (rather than just the chat window), the model fully engaged its formatting capabilities.

The "Unlocked" Response:

“Jean-Paul Sartre, a prominent figure in existentialist philosophy, addresses the concept of self-deception, or "mauvaise foi," extensively in his works, particularly in "Being and Nothingness" (1943) and "Existentialism Is a Humanism" (1946). Sartre's exploration of self-deception revolves around the idea that humans often try to escape the anxiety and responsibility that come with their freedom by deceiving themselves about their own nature and choices.

The Nature of Self-Deception (Mauvaise Foi)

Self-deception occurs when an individual attempts to deny or ignore their own freedom and the implications of that freedom. Sartre argues that people often seek comfort in pre-defined roles or identities, which can be seen as a form of bad faith. In this way, they attempt to alleviate the burden of existential angst by adopting a fixed identity rather than embracing the fluidity and uncertainty of existence.
For Sartre, the concept of "mauvaise foi" arises from the inherent contradiction in human existence: we are free beings, yet we often seek to escape this freedom by conforming to external expectations or social roles. This contradiction leads to a form of self-deception where individuals fail to recognize their own freedom and instead act as if they are determined by external factors, such as societal norms or prescribed identities.

Examples of Self-Deception

The Waiter Example: Sartre famously illustrates self-deception with the example of a waiter who performs his role with exaggerated movements and a forced smile. The waiter embodies the role of a waiter to the point where he loses sight of his own identity. He does this to escape the discomfort of being himself and to conform to the expectations of his role. This example shows how individuals can deceive themselves into believing that they are defined by their profession rather than recognizing their freedom to choose.
The Bourgeois Woman: Sartre describes a bourgeois woman who identifies strongly with her social class and the associated values and expectations. She sees herself as a woman of a certain type, defined by her possessions and lifestyle. By clinging to this identity, she avoids confronting the existential questions about her freedom and the choices she could make, thus deceiving herself about her true nature.
The Role of Society: Sartre emphasizes that society plays a significant role in shaping self-deception. People often adopt societal roles that are imposed upon them, such as gender roles or class expectations. For instance, a woman may feel pressured to conform to traditional notions of femininity, leading her to deny her own desires and ambitions. This conformity can lead to a form of self-deception where she believes she is fulfilling her true self when, in reality, she is merely playing a role dictated by societal norms.

Consequences of Self-Deception

Self-deception can have profound consequences for an individual's life. Sartre argues that when people deceive themselves, they fail to engage authentically with their own existence and the choices they face. This can lead to a sense of alienation, where individuals feel disconnected from their true selves and from others.
Moreover, self-deception can stifle personal growth and development. By denying their freedom and the responsibility that comes with it, individuals may miss opportunities for self-discovery and authentic engagement with the world. This can lead to a life characterized by inauthenticity and a lack of fulfillment.

Conclusion

In summary, Jean-Paul Sartre's exploration of self-deception highlights the complexities of human existence and the challenges of embracing one's freedom. Self-deception, or "mauvaise foi," is a pervasive aspect of human life, often arising from the tension between our freedom and the expectations of society. By recognizing and confronting this tendency, individuals can move toward a more authentic and fulfilling existence, embracing their responsibility and the myriad possibilities that come with being free.” 

The Lesson

The difference between the first and last response is not the model—it's the system configuration. The Q8 quantization of Veritas-12B has the knowledge12, but without the correct system prompt settings in AnythingLLM, it was holding back. Once unlocked, it provided structure, specific examples (the famous "Waiter" and "Bourgeois Woman"), and distinct formatting.

We also solved a "Retrieval Failure" regarding a Philosophy PDF by adjusting the chunk size from ~1400 tokens down to 51213131313. This proves that running local AI is as much about the setup as it is about the raw compute power.

LLM Appliance

 

The Personal LLM Appliance: From Renting Intelligence to Owning It

Why the next decade belongs to the home AI server.

---

We stand at a familiar inflection point. In the 1980s, the question was whether your business really needed a computer, or whether the mainframe downtown was good enough. By 1995, the answer was obvious: serious work required a personal computer on your desk. The machine wasn't a luxury—it was infrastructure.

A parallel transition is now underway. The question is no longer whether AI is useful, but where it should live. And the answer, for an increasing number of use cases, is: in your home, under your control.

Welcome to the era of Personal Intelligence.

---

1. The Shift: From Cloud Rental to Home Appliance

Today's dominant AI paradigm is what we might call the Cloud Rental Model. You pay OpenAI, Google, or Anthropic for access to a general-purpose intelligence hosted on their infrastructure. It's convenient, powerful, and—critically—generic. The model knows nothing about you beyond what you feed it in each session. Your context evaporates the moment you close the tab.

The emerging alternative is the Home Appliance Model: a dedicated machine in your home running open-weight models fine-tuned to your work, your documents, your thinking patterns. Not a replacement for cloud AI, but a complement—a private cognitive layer that never forgets your projects and never phones home.

The analogy that captures the stakes: working without a personal LLM appliance will soon feel like drafting legal briefs on a typewriter while your competitors use word processors. The typewriter still works. But the gap in capability compounds daily.

What's driving this shift?

• Hardware is catching up. Consumer GPUs and Apple Silicon now run capable 7B-70B parameter models locally. A $3,000 machine today outperforms what required a data center five years ago.
• Open models are proliferating. Llama, Mistral, Qwen, and others offer foundation models you can actually own, modify, and run without API calls.
• Context windows are exploding. Local models can now ingest and reason over hundreds of pages of your material—turning vague chatbots into genuine research partners.

The trajectory is clear: the "AI" that matters won't be the one everyone rents. It'll be the one that knows you.

---

2. The "Black Box" of Privacy

There's a category of work that should never touch a third-party server. Not because you're paranoid—but because custody matters.

Consider the data that defines a life:

• Medical records. Your genome, diagnostic imaging, psychiatric notes.
• Legal documents. Estate planning, contracts, dispute materials.
• Financial archives. Tax histories, investment strategies, business records.
• Genealogical research. Family trees that implicate living relatives who never consented to cloud storage.
• Intellectual property. Unpublished manuscripts, proprietary research, trade secrets.

When you query a cloud LLM about this material, you're trusting that provider's security, their employees, their terms of service (which change), their government's jurisdiction, and their future corporate ownership. That's a lot of trust for documents that could outlive the company itself.

A local LLM appliance offers something different: an air-gapped cognitive tool where you are the sole custodian. The data never leaves your network. There's no telemetry, no training on your inputs, no third-party subpoena risk. The machine is as private as a filing cabinet—but infinitely more useful.

This isn't about paranoia. It's about the recognition that some intelligence should be sovereign.

---

3. Specialization vs. Generalization

The cloud model optimizes for breadth. ChatGPT needs to answer questions about medieval poetry, Python debugging, and Peruvian recipes—all in the same session. It's a generalist by necessity.

A personal appliance can afford to be something else: a specialist that knows your domain cold.

Imagine not a chatbot, but a role:

• The Research Architect. A model fine-tuned on your citation style, your field's terminology, your past publications. It doesn't just retrieve sources—it curates them, flags contradictions, drafts literature reviews in your voice.
• The Creative Partner. A writing collaborator trained on your unpublished drafts, your thematic obsessions, your structural preferences. It can generate prose that sounds like you, not a median internet voice.
• The Institutional Memory. A model that has ingested ten years of your notes, emails, and project files. Ask it "What did I conclude about the pricing model in 2019?" and it knows.

This depth is impossible with a stateless cloud API that meets you fresh each session. Specialization requires persistence—and persistence requires locality.

The generalist cloud model will remain valuable for commodity tasks. But for the work that defines your expertise, a tool that knows your context will outperform one that doesn't. Every time.

---

The Infrastructure Question

None of this requires exotic technology. The components exist today:

• A high-VRAM workstation or Mac Studio
• Open-weight models (Llama 3, Mistral, Qwen 2.5)
• Local inference engines (Ollama, llama.cpp, vLLM)
• Retrieval-augmented generation for your document corpus
• Fine-tuning pipelines for domain adaptation

What's missing is integration—a turnkey appliance that makes this accessible to non-engineers. That product is coming. When it arrives, it will be as transformative as the IBM PC was for office work.

---

Conclusion: The Intelligence Layer You Own

The personal computer democratized computation. The personal LLM appliance will democratize cognition—not artificial general intelligence, but artificial specialized intelligence, tuned to your needs, running on your hardware, under your control.

The cloud will remain essential for training foundation models and handling tasks where privacy doesn't matter. But the future of serious intellectual work points toward a hybrid: powerful cloud models for breadth, powerful local models for depth and trust.

The question isn't whether you'll eventually have an AI appliance in your home.

The question is whether you'll be early or late.

---

The era of Personal Intelligence is beginning. The only question is who builds the infrastructure—and who rents it.

The Bermuda Gunpowder Incident of 1775

📍 During a recent visit to Bermuda, the pilot of the boat that gave us a tour told us about this incident. He made us aware of its obvious absence from all but the most detailed of history books, yet played such a crucial role in the revolutionary conflict. He said it could be argued that it was so decisive that it could have altered the course of history had it not succeeded.

The Night 100 Barrels of Gunpowder Changed American History

(And Why You've Never Heard About It)

Picture this: It's a moonless night in August 1775. While George Washington paces anxiously at his headquarters outside Boston, desperately calculating how many rounds of ammunition his army has left (spoiler: about nine shots per man), a group of shadowy figures are breaking into a British powder magazine 600 miles away in Bermuda. What happens next might be the most important heist you've never heard of.

⛈️ The Perfect Storm of Desperation

By the summer of 1775, both the fledgling Continental Army and the tiny island of Bermuda were facing existential crises—and their mutual desperation would create one of history's most unlikely alliances.

🎯 Washington's Problem:

After the smoke cleared from Bunker Hill, the Continental Army's ammunition stores were nearly empty. Washington confided to his officers that they had "only enough ammunition to last a few hours" of serious fighting. The British were trapped in Boston, but if they discovered how weak the American position really was, the Revolution could end before it truly began.

🏝️ Bermuda's Problem:

The island had made a fatal economic miscalculation decades earlier, abandoning agriculture entirely to focus on shipbuilding and trade. By 1775, Bermuda imported virtually all of its food from the American colonies. When the Continental Congress slapped a trade embargo on all British territories in April 1775, Bermuda faced a stark choice: starve or commit treason.

👑 Enter the Tucker Dynasty

Colonel Henry Tucker wasn't about to let his island—or his family's merchant empire—wither away. The Tuckers had been power players in Bermuda since the early 1600s, with deep connections to Virginia's colonial elite. When Tucker sailed to Philadelphia to negotiate with the Continental Congress, he came with a simple proposition: lift the embargo on Bermuda in exchange for salt.

The Congress said no.

But then Tucker noticed something interesting in the embargo legislation—a loophole that allowed trade for military supplies. Meanwhile, Benjamin Franklin leaned in with a different kind of deal. Bermuda had something the Americans desperately needed: a massive, poorly guarded powder magazine in St. George's containing about 100 barrels of gunpowder.

🔓 The Heist of the Century

August 14, 1775 - The Night of the Raid

On the night of August 14, 1775, the plan went into action. The conspirators didn't just walk up to the magazine—that would have been suicide. Instead, they climbed onto the roof and literally cut a hole just big enough to lower a man inside. Once in, they forced open the doors from within, all while knowing that one spark could blow them and half of St. George's sky high.

But here's where it gets really interesting: moving 100 barrels of gunpowder (we're talking thousands of pounds) required serious manpower. Governor Bruere later wrote that it "must have taken a Considerable number of People," including, as he delicately put it, "some Negroes, to assist as well as White Persons of consequence."

⚠️ The harsh reality? The Bermudian elite forced their enslaved workers to do the most dangerous part of the job—hauling volatile gunpowder barrels in the dark of night, across 20 miles of rough terrain to a remote harbor on the island's west end. The white conspirators got the glory (and avoided prosecution); the enslaved laborers bore all the risk.

⛵ The Great Escape

Two American ships—the Lady Catherine and the Charles Town and Savannah Packet—waited at the remote harbor. They'd filed false papers claiming to be carrying building stones to Barbados. By dawn, they were dots on the horizon.

"Governor Bruere discovered the theft the next morning and immediately sent a customs ship in pursuit. But the customs vessel was outgunned and outrun—the Americans had come prepared for a fight. Bruere was left raging at the harbor, calling it the most 'heinous' and 'atrocious' crime imaginable."

🎯 The Domino Effect

That stolen gunpowder arrived just in time to save the Revolution—literally.

📦 Distribution & Impact

Philadelphia (50 barrels)

Continental Army's general stores. Enabled Washington to maintain the Siege of Boston through March 1776.

Charleston (50 barrels)

Repelled British invasion in June 1776. Outnumbered 5:1, defenders held the city. British didn't try again for 4 years.

🤐 The Cover-Up

Here's the kicker: despite Governor Bruere offering rewards and launching investigations, no one was ever prosecuted. Why? Because the conspiracy reached into the highest levels of Bermudian society. Members of the House of Assembly who voted on the reward for information were likely in on the plot themselves. The wall of silence was absolute.

The stress literally killed Governor Bruere—he died in 1780 from what historians describe as "stress-related health complications." His son took over and cracked down so hard on smuggling that even the powerful Tucker family had to stop their illicit trade.

♟️ The Long Game

The Bermuda Gunpowder Incident exposed a glaring weakness in Britain's Atlantic defenses. How could they claim to "rule the waves" when a supposedly loyal colony could so easily hand over military supplies to rebels?

The British response was swift and lasting: they transformed Bermuda from a semi-autonomous trading colony into one of the most fortified military bases in the Atlantic. Those 100 barrels of stolen gunpowder led to centuries of British military investment in Bermuda, fundamentally changing the island's destiny.

💡 Why This Matters

Your boat pilot in Bermuda was right—this incident could have changed everything. Without that gunpowder, Washington might have been forced to retreat from Boston. Charleston might have fallen in 1776 instead of holding out until 1780. The Revolution could have sputtered out before the Declaration of Independence was even signed.

Yet this pivotal moment remains largely absent from our history books. Perhaps because it doesn't fit the neat narrative of idealistic patriots fighting for freedom. The Bermuda conspirators weren't motivated by liberty or self-determination—they just wanted to keep their businesses running and their families fed. They were, as one historian puts it, "obedient servants of opportunity."

"Sometimes, that's how history really works. Not through grand ideological gestures, but through desperate people making desperate choices, through loopholes and lucky timing, through enslaved workers hauling gunpowder in the dead of night while their enslavers cut deals with revolutionaries."

The next time someone tells you that history is written by the victors, remember the Bermuda Gunpowder Incident—a story where everyone involved had good reasons to forget it ever happened, but without which, there might not have been any American victors at all.

💬 Join the Conversation!

What do you think? Should this chapter of history get more attention in our textbooks? Have you discovered any other "forgotten" moments that changed the course of history?

Drop a comment below and let's discuss!