Build Andrej Karpathy’s LLM Knowledge Base for Businesses (10x Output!)
This video walks through setting up Andrej Karpathy's 'LLM wiki' second-brain system for a business by creating an Obsidian vault, pasting in Karpathy's GitHub-gist idea file, and letting Claude Code build a raw/schema/wiki structure that ingests your business data into a self-linking, queryable knowledge base.
Dream Labs AIWatchTranscript found
Quick learning frame
Read this before watching.
Coding-agent workflow is the loop of inspect, route, plan, edit, verify, summarize, and decide what should be automated next.
New playlist item from Dream Labs AI; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: Building a personalized AI knowledge base for a business: structuring raw business data, applying Karpathy's schema/rulebook, and using Claude Code plus Obsidian to compile and query a self-updating second brain.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01Inspect context
02Route tool
03Plan work
04Edit safely
05Verify behavior
06Report next step
Deep lesson
Turn this video into working knowledge.
3,731 cleaned transcript words reviewed across 1,038 timed caption segments.
Thesis
Build Andrej Karpathy’s LLM Knowledge Base for Businesses (10x Output!) teaches a practical coding-agent workflow move: This video walks through setting up Andrej Karpathy's 'LLM wiki' second-brain system for a business by creating an Obsidian vault, pasting in Karpathy's GitHub-gist idea file, and letting Claude Code build a raw/schema/wiki structure that ingests your business data into a self-linking, queryable knowledge base.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:16
Generalist vs specialist AI
“real context, and no alignment or automation towards their actual business goal. So, Andre created a game-changing system called your AI second brain. This system is so powerful that even his highly technical tweet about it reached over...”
Most people query an LLM directly against the whole internet and get average 'AI slop'; Karpathy's fix is to filter prompts through a personal layer (goals, business context, rules, history) so the model's output is specific to your business rather than the average of all training data. Write down 3-5 business outputs you currently ask AI for (emails, content, planning) and note what business context the model is missing each time, to define what your filter layer must contain.
10:26
Compile the wiki
“drag and drop this straight in. You can actually have like a Rolls Model skills plugin, which is what I'm personally working on for my business. So, you take all their publicly available data, and you say maybe...”
After dropping raw assets (goals, transcripts, competitor notes, interview answers, PDFs) into the raw folder, you tell Claude Code to compile; it reads sources in parallel and writes source summaries, entity pages, and concept pages, synthesizing data once into an AI-native structure that is fast to query later. Assemble a raw-data folder for your own business (goals file, a few transcripts/docs, competitor notes) and run the compile step, watching which source, entity, and concept pages Claude generates.
11:54
Self-learning log
“practical, not novelty. What's actually moving me and my business forward? Every tool, framework, agent, idea we build for recommended must actually drive results, revenue, time saved, audience grown, or decisions made. If it's clever but doesn't move...”
A log file records every interaction with the vault, and Claude updates its own memory over time; this is the mechanism that turns a flat knowledge base into compounding, business-specific knowledge that gets smarter with each query and piece of feedback. After compiling, run a few real queries and give feedback, then open the log and index files to confirm the loop is capturing interactions and updating the vault.
01
Inspect context
Start with this video's job: This video walks through setting up Andrej Karpathy's 'LLM wiki' second-brain system for a business by creating an Obsidian vault, pasting in Karpathy's GitHub-gist idea file, and letting Claude Code build a raw/schema/wiki structure that ingests your business data into a self-linking, queryable knowledge base. Treat "Inspect context" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:16, where the video says: “real context, and no alignment or automation towards their actual business goal. So, Andre created a game-changing system called your AI second brain. This system is so powerful that even his highly technical tweet about it reached over...”
02
Route tool
Use "Route tool" to locate the part of the coding-agent workflow mechanism the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 10:26, where the video says: “drag and drop this straight in. You can actually have like a Rolls Model skills plugin, which is what I'm personally working on for my business. So, you take all their publicly available data, and you say maybe...”
03
Plan work
Turn "Plan work" into the reusable artifact for this lesson: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal. This is where watching becomes something you can inspect and reuse.
04
Edit safely
Use "Edit safely" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Verify behavior
Use "Verify behavior" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Report next step
Use "Report next step" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed artifact packet
Convert the video into a scoped artifact request that includes the transcript claim, mechanism, acceptance criteria, and proof. The output should be a coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal..
Example
Coding-agent workflow proof brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the coding-agent workflow pattern.
Example
Teach-back module
Transform the lesson into a definition, a Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
choosing tools by hype
losing context across agents
letting parallel sessions become invisible
Letting the lesson drift into generic Codex vs Claude comparison.
Letting the lesson drift into feature lists without task routing.
Letting the lesson drift into claims that ignore limits or recovery.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: This video walks through setting up Andrej Karpathy's 'LLM wiki' second-brain system for a business by creating an Obsidian vault, pasting in Karpathy's GitHub-gist idea file, and letting Claude Code build a raw/schema/wiki structure that ingests your business data into a self-linking, queryable knowledge base.
02
Explain the practical stakes without hype: New playlist item from Dream Labs AI; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: Build Andrej Karpathy’s LLM Knowledge Base for Businesses (10x Output!)
- URL: https://www.youtube.com/watch?v=FAWm7DuFSPc
- Topic: AI Strategy
- My current learning frame: Build your own business second brain: create an Obsidian vault, paste Karpathy's LLM-wiki idea file, drop in at least one goals document plus three real business assets, and use Claude Code to compile it, then query it for where the gaps in your business knowledge are.
- Why this matters: New playlist item from Dream Labs AI; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:16 / Evidence 1: "real context, and no alignment or automation towards their actual business goal. So, Andre created a game-changing system called your AI second brain. This system is so powerful that even his highly technical tweet about it reached over..."
- 2:47 / Evidence 2: "self-learning system. They're only going to get better, smarter, and more catered to you and your business needs in the future, because every time you make a query, every time you need something done, every time you give..."
- 4:37 / Evidence 3: "Obsidian for macOS, or if you're on PC, grab it for PC. And of course, you're going to need to download Claude or Claude code. I'm personally going to be using it over here in the terminal, but..."
- 7:15 / Evidence 4: "this is where Claude Code will do all of the heavy lifting for us and it's going to go and set it up. So, it says, "Got it. Let me build out the LLM wiki pattern in your..."
- 10:26 / Evidence 5: "drag and drop this straight in. You can actually have like a Rolls Model skills plugin, which is what I'm personally working on for my business. So, you take all their publicly available data, and you say maybe..."
- 11:54 / Evidence 6: "practical, not novelty. What's actually moving me and my business forward? Every tool, framework, agent, idea we build for recommended must actually drive results, revenue, time saved, audience grown, or decisions made. If it's clever but doesn't move..."
- 16:48 / Evidence 7: "Logline, "Take the exact Hormozi AI stack, 14 prompts, five agents, firing at 6:00 a.m., and document what 30 days actually produce. Real numbers, real wins, real disasters." Why this one? It passes every editorial filter. I ask..."
Video-aware target:
- Prompt lane: Coding-agent workflow
- Mechanism to extract: Find the workflow rule that explains when and how to use Codex, Claude Code, browser control, dashboards, or manual review.
- Artifact to produce: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
- Artifact must include: task class; agent/tool choice; context packet; verification step; handoff/recovery rule
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Find the workflow rule that explains when and how to use Codex, Claude Code, browser control, dashboards, or manual review. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step
- answers to these source questions: What workflow pain is being solved? | What exact coordination mechanism is shown? | What evidence proves it changes the work?
- 3 concrete examples that apply the video idea to real agentic work, such as rate-limit routing; browser verification after a UI edit; long-running /goal session review
- 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: choosing tools by hype; losing context across agents; letting parallel sessions become invisible
- a checklist for the next real workflow, focused on: routing decision, context portability, verification, handoff summary
- one practical exercise with a clear done signal: Route three recent tasks across Codex, Claude, browser checks, and manual review with a reason for each.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "Build Andrej Karpathy’s LLM Knowledge Base for Businesses (10x Output!)", not a generic AI Strategy essay.
- Each workflow rule must point to a timestamped claim or demo moment, then state what remains unproven.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic Codex vs Claude comparison; feature lists without task routing; claims that ignore limits or recovery.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
Every new AI tool deserves a trial.
Every tool has integration cost. Start from workflow pain, not novelty.
If an agent can do it once, it is automated.
Automation means repeatable, monitored, recoverable, and reviewable.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal..
A reusable artifact with a done signal and one verification step.03
Coding-agent workflow teach-back card
Explain the coding-agent workflow mechanism to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
According to the video, what concretely changes between the 'generalist' way most people use an LLM and Karpathy's 'specialist' approach?
Once you've dropped your raw assets into the raw folder and tell Claude Code to compile the wiki, what specific kinds of pages does it write, and why structure it that way?
What is the role of the log file in Karpathy's system, and why is it called the self-learning mechanism?
Source shelf
Use the video as a doorway, then verify with primary sources.
