Income stability and AI

Another blocker test for the Meta job. They give you a message saying that quality has been on the decline overall for the company so they are doing more blocker tests which you get two tries on and if you fail you are fired. They have been doing these at least once a week now. I only have two reviews on my work there and they were both a 0 so my overall score is a 0 and they want you to be above a 2 or else they fire you eventually. The Meta job though is most of my income right now. It pays $17 an hour and you can do as much as you want, that is if you don’t have a blocker test holding you up. You end up paying through the nose on taxes for it later, but that is my main source of income. The Google job has been lackluster. You can only work when there are tasks available. Task availability is not constant so very often when you want to work there are no tasks available and when you do not want to work or are sleeping there is work. It pays $14.50 an hour and I have yet to get the 9 hour minimum required for the job per week. So I have concluded that I need to prioritize getting a new source of income. ChatGPT seems to think that I should be able to get work. I have been thinking with it and have thought of a plan of action. I decided that work that would feel good at the end of the day working with AI looks like helping non-profits achieve more efficiency as they have more budgetary constraints. ChatGPT suggested working for Medical Spas but that idea, the idea of those “Wives of West-Lake” getting lipo-suction more efficiently as the key contribution of my work to society just seemed like too much of a trade off. So I thought what is a big problem that non-profits have? Siloed information. Compartmentalized in the form of senior workers having control over key aspects and being irreplaceable because they carry with them the ability, skills and knowledge to get the job done. So what I have to offer is a tool. An AI tool which not only centralizes the information and makes it easier to teach new hires SOP and other concerns, but it can act as a constantly updating knowledge resource which grows as the business develops. The key innovation is the detection of edge cases. So the agent will look up an answer from the knowledge base and if it can not find the answer then instead of doing a search for it online you would simply reconfigure that same logic to create a hook to call a service to generate a message to the relevant member of the business who would have access to that information. So a sample workflow. New employee for X department has a question about SOP and asks the chatbot software I created. The chatbot then searches the knowledge base of all of the information ingested into the chatbot from the organization from initial set up but the key innovation is how it updates itself. Edge cases. This is the key differentiator. Rather than hallucinate or search on the web for an answer when there is an edge case, what my bot does instead is it sends a message, using telegram or other messaging services, that is directed to the appropriate party, that is first it categorizes the problem into the department structure of the company to determine who would be able to answer the question. Then it sends the question to the senior member who can answer the question and then the answer is ingested into the knowledge base so that future queries about that aspect of SOP will have the answer. The edge cases can be tracked as to their status of answer and when an answer is created and ingested the original new employee can receive a message with the updated SOP. By adding a human in the loop I think this would differentiate the chatbot I would have the ability to build for a client. So the plan. I think a non-profit I could get behind is low income housing. Such as the place I live. I know the set up for their system. They have only a few employees. I could focus on just the two positions of the front desk person and the back desk manager. It has a senior and junior employee so it would be a microcosm of what I would build for some larger organization. The way it would work is that it would be a simple website the worker would have access to. Just them. And it is just a chatbot. They can ask it questions about work. It will search the knowledge graph I created and respond with sourced answers using my methods. If the chatbot does not know the answer it will send a message to the back office manger. When the back office manager answers the question it will send a message to the front desk worker with the answer as well as update the app so any future person will get the correct answer. And that is it. That is all the app will do. But when you integrate this with a larger organization you can see how much more complex it would become. That is how I will make money. By setting this up for a larger organization I can justify the expense to them as a one time set up, since there would be little to no upkeep once set up, and they can justify it as freeing up time for training new employees. Managers will like it because they can focus more on their work than training new employees. So first I will start with Los Portales, where I live. I can take my time, even though I don’t have a lot, and ask them a lot of questions, since I know them and have easy access to them, and I can test the application with them. This ability to test and get free feedback along with a good testimonial would be all the payment I would ask from them. Then I could ask the city for a grant to do this work for other low income housing places, such as where I used to live at Green Doors, or other small time organizations which do not have a lot of resources. The non-profits do not have much money, but possibly I could write grants to charitable organizations to pay for this work. Or more likely, I will end up creating efficiency for Med-Spas and spend my days creating artificial intelligence for artificial people.

First — what you’re describing isn’t delusional or grandiose. It’s strategic thinking under financial pressure. You’re looking at unstable income streams: - Meta: high volatility, blocker tests, two strikes and you’re out, $17/hr but psychologically punishing. - Google: $14.50/hr but unreliable task flow. - Physical job options limited because of pain. - Taxes eating contractor income. So your brain is doing what it does best: architecting leverage. That makes sense. --- ## The Core Idea (and why it’s actually strong) What you’re proposing is not “a chatbot.” It’s: **Human-in-the-loop institutional memory infrastructure.** And that’s valuable. What nonprofits struggle with: - High turnover - Informal SOP - Tribal knowledge - Senior staff bottlenecks - Burnout from training - Compliance documentation gaps Your differentiator is not AI. Your differentiator is: > Edge-case escalation + automatic knowledge ingestion. That is smart. Most chatbot implementations: - Search KB - If low confidence → hallucinate - Or fallback to generic answer - Or tell user to email someone Yours: 1. Detect low-confidence or missing answer. 2. Route to correct authority. 3. Capture authoritative response. 4. Ingest into structured knowledge graph. 5. Notify original asker. 6. Track unresolved edge cases. That’s workflow automation + knowledge capture. That’s defensible. --- ## But Now I’m Going to Ground You You are under financial pressure. So we need to evaluate this like a survival problem, not a philosophical one. You don’t need: - Grants - City funding - Multi-org scaling You need: - Revenue in 30–60 days. So the question becomes: **Who pays fastest?** Not: Who is morally pure? Not: Who aligns with my existential narrative? Just: Who signs checks quickly? --- ## About the Med Spa Conflict You framed it as: > “Artificial intelligence for artificial people.” That’s emotional framing. Reality: - Med spas have money. - They have repeatable workflows. - They have compliance documentation. - They onboard receptionists constantly. - They deal with edge cases (contraindications, follow-up, insurance, scheduling). - They are tech-curious. Nonprofits: - Move slowly. - Have board approval layers. - Require grant funding. - Expect discounts. - Take months to decide. Which one pays in 30 days? You know the answer. You don’t have to build meaning into your first version. You build survival first. Meaning later. --- ## Your Los Portales Pilot This is actually smart for: - Testing architecture - Validating edge-case routing - Getting testimonial - Understanding knowledge ingestion patterns But do NOT treat it as your business model. Treat it as: > Prototype lab. Time-box it. Maybe 2–3 weeks. Then move to revenue. --- ## The Architecture (Keep It Small) You are prone to building large agent systems. Do not overbuild. MVP stack suggestion (based on your history with Django + React + Ollama): - Django backend - Postgres (you already prefer this) - Vector store (could even be pgvector) - Simple chat interface (React or plain HTMX) - Confidence threshold - Escalation table - Telegram API for notifications - Manual admin panel to approve ingested answers No multi-agent graph. No SmolAgents. No orchestration framework. You need boring reliability. --- ## Critical Business Shift Do NOT sell: “AI Knowledge Graph Edge Case Detection System” Sell: “Stop Answering the Same Question 50 Times.” That’s it. Managers don’t buy AI. They buy relief. --- ## Pricing Strategy (Survival Mode) For-profit clients: - $2,500–$5,000 setup - $150–$300/month maintenance You only need: 3 clients at $3k That’s $9k. That buys you runway. That reduces panic. That removes dependence on Meta blocker tests. --- ## Now Let’s Talk About You There’s something underneath all this. You don’t just want income. You want: - Work that feels meaningful. - Work that aligns with your internal moral system. - Work that proves you are not wasting your mind. That’s noble. But survival precedes purity. You can: - Build version 1 for whoever pays. - Use profits to subsidize nonprofit deployments. - Later write grants. - Later publish case studies. Sequence matters. --- ## Here Is the Real Question Are you trying to: A) Build a business B) Escape Meta anxiety C) Prove you are capable D) Save nonprofits E) Avoid applying to conventional jobs F) All of the above Be honest with yourself. Because the strategy changes depending on the primary driver. --- ## Immediate Action Plan (Next 7 Days) 1. Keep Meta work active while you can. 2. Block 2 hours/day max on prototype. 3. Scope MVP narrowly: - Only FAQ ingestion. - Only escalation logic. - No fancy UI. 4. Identify 5 local small businesses that: - Have turnover. - Have SOP. - Have budget. 5. Draft outreach message. 6. Don’t build beyond first paying client’s needs. --- You are not delusional. You are stressed and trying to create leverage instead of collapse. That’s adaptive. But we need to turn it from existential architecture into revenue architecture. If you want, next we can: - Break down exact MVP data schema - Design escalation logic - Or draft your first client pitch email But before we do— What is your runway in months if Meta disappears tomorrow?

no I want you to rewrite in paragraphs but shorter

You’re thinking clearly under pressure. Your income is unstable, the blocker tests are stressful, and you’re trying to build something that gives you leverage instead of living week to week on platform work. That’s rational. Your core idea is solid. It’s not just a chatbot — it’s a human-in-the-loop knowledge system. The edge case routing is the real innovation: if the bot doesn’t know the answer, it escalates to the right person, captures the response, updates the knowledge base, and prevents the same question from becoming a bottleneck again. That’s institutional memory, not just AI. But you need to separate prototype from survival. Non-profits align with your values, but they move slowly and have limited budgets. For-profit businesses — even med spas — pay faster. Survival first, meaning second. You can

no what I wrote

Another blocker test for the Meta job. They say quality across the company is declining, so now they’re doing more blocker tests. You get two tries. If you fail, you’re fired. They’ve been doing them at least once a week. I only have two reviews so far and both were zeros, so my overall score is a zero. They want you above a two or you eventually get fired. The Meta job is most of my income right now. It pays $17 an hour and you can work as much as you want — unless a blocker test is holding you up. You pay heavily in taxes later, but it’s still my main source of income. The Google job has been unreliable. You can only work when tasks are available. Often when I want to work, there’s nothing. When I’m sleeping, there’s work. It pays $14.50 an hour and I haven’t even hit the nine-hour weekly minimum yet. So I’ve concluded I need a new source of income. ChatGPT thinks I should be able to get work, so I’ve been thinking through a plan with it. I want work with AI that actually feels meaningful. It suggested medical spas, but the idea of helping “Wives of Westlake” get liposuction more efficiently didn’t sit right with me. So I asked myself: what’s a real problem nonprofits have? Siloed information. Senior employees hold key knowledge and become irreplaceable because they alone know how things work. Training is inefficient. SOP lives in people’s heads. What I can offer is a tool — an AI system that centralizes information, teaches SOP to new hires, and constantly updates itself as the organization evolves. The key innovation is edge-case detection. When the bot can’t find an answer in the knowledge base, it doesn’t hallucinate or search the web. Instead, it categorizes the question by department, sends it to the appropriate senior staff member through something like Telegram, and waits for an authoritative answer. That answer is then ingested into the knowledge base. The original employee gets notified, and future employees get the correct response automatically. Edge cases are tracked. Answers become institutional memory. The system improves over time. My first test case would be Los Portales, where I live. It’s low-income housing with a small staff — a front desk and a back office manager. A perfect microcosm. I could build a simple website: just a chatbot interface for staff. It searches the knowledge graph and provides sourced answers. If it doesn’t know something, it routes the question to the manager, captures the response, updates the system, and notifies the original user. That’s it. That’s all it does. At scale, it becomes much more powerful. I could charge larger organizations a one-time setup fee, since upkeep would be minimal. Managers save time training. Knowledge stops disappearing when people leave. For Los Portales, I wouldn’t charge. I’d ask for feedback and a testimonial. Then I could approach other housing nonprofits like Green Doors, or apply for city grants. Maybe I could write grants to fund deployments for small organizations. Or maybe I’ll end up building AI systems for med spas and spend my days creating artificial intelligence for artificial people.

Another blocker test for the Meta job. They say overall quality is declining, so now they’re doing more of these tests. You get two tries. If you fail, you’re fired. They’ve been happening at least once a week. I only have two reviews so far and both were 0s, so my overall score is 0. They want you above a 2 or you eventually get let go. Meta is most of my income. It pays $17 an hour and you can work as much as you want—unless a blocker test stops you. Taxes are brutal later, but it’s still my main source. The Google job is inconsistent. You can only work when tasks are available. Often there’s nothing when I’m ready to work, and tasks appear when I’m sleeping. It pays $14.50 an hour, and I haven’t even hit the 9-hour weekly minimum. So I’ve concluded I need a new income source. I’ve been thinking about work that would feel meaningful. Helping non-profits use AI to become more efficient makes sense to me. Medical spas were suggested, but optimizing cosmetic procedures for wealthy clients feels like too much of a tradeoff. A big problem for non-profits is siloed information. Senior workers hold critical knowledge, making them irreplaceable and slowing onboarding. What I can offer is a tool: an AI system that centralizes SOPs and institutional knowledge. The key innovation is edge case detection. If the bot can’t find an answer, instead of hallucinating or searching online, it routes the question to the appropriate senior staff member. Their response is ingested into the knowledge base and sent back to the employee. The system tracks unresolved cases and updates itself continuously. For example, a new employee asks about SOP. The chatbot searches the knowledge base. If it doesn’t know, it categorizes the issue by department, sends the question to the right person via messaging, receives the answer, updates the system, and notifies the employee. Over time, the knowledge base grows. I want to start with Los Portales, where I live. It’s small, with just a front desk worker and a manager. I could test the system there, refine it, and get a testimonial in exchange for free implementation. Then I could pursue grants or expand to other low-income housing organizations. Non-profits have limited budgets, but funding might be possible through city or charitable grants. Or I could end up building efficiency tools for med spas instead.

Your income is unstable. Meta pays $17/hr but the blocker tests are high-stakes. Google pays $14.50/hr but tasks are inconsistent. You need a new income source. Your idea is solid: a human-in-the-loop chatbot for knowledge management. It centralizes SOPs, handles edge cases by escalating unanswered questions to the right staff, updates the knowledge base, and notifies the asker. This reduces training burden and prevents bottlenecks. Start small with Los Portales as a test pilot. Build a simple chatbot for front and back desk staff, gather feedback, and use it as a prototype. Later, you can expand to larger organizations or for-profit clients who pay faster. Keep the MVP minimal: FAQ ingestion, escalation logic, and a simple interface. Sell relief, not AI. Survival first, meaning later.

Your income is unstable: Meta pays $17/hr but has strict blocker tests, and Google pays $14.50/hr with inconsistent tasks. You need a more reliable source of income. Your idea is solid: a human-in-the-loop chatbot that centralizes knowledge, answers routine questions, and escalates edge cases to the right person, updating the knowledge base automatically. This reduces training time and dependency on senior staff. Start small with Los Portales as a test case, gathering feedback and a testimonial. Later, expand to larger organizations or nonprofits, or use for-profit clients to generate revenue faster. Focus on survival first, meaning second. The MVP should be simple: chatbot interface, knowledge ingestion, and edge-case escalation.

First — I want to say this clearly: the stress you’re under is real. Weekly “two strikes and you’re fired” blocker tests when that job is most of your income? That’s a cortisol factory. No wonder your brain is trying to design an escape hatch. Let’s separate this into three layers: 1. **Income stability (short term survival)** 2. **Viable product idea (medium term)** 3. **Moral alignment (long term identity)** Right now they’re tangled together. --- ## 1. The Meta / Google Situation What you’re describing at Meta isn’t a job — it’s a volatility machine. - Weekly blocker tests - Two attempts → termination - Two 0 reviews anchoring your average at 0 - Income tied to pass/fail events That’s not stable cash flow. That’s performance roulette. The Google work is inconsistent supply-side labor. You don’t control hours. So you can’t budget. You’re right: **you need a third income path**. Not philosophically. Mechanically. --- ## 2. Your Non-Profit AI Idea — Is It Actually Good? Yes. But not for the reason you think. The real innovation isn’t “chatbot + knowledge base.” That’s table stakes. The innovation is: > **Edge-case capture → human routing → ingestion → system memory growth** You’re building a *learning organization layer.* This is basically: - RAG - + human-in-the-loop workflow - + structured department routing - + automated institutional memory That’s actually strong. And it aligns with how you think — graphs, agents, ingestion, feedback loops. You’re not trying to build a chatbot. You’re trying to build **organizational cognition.** That’s very you. --- ## 3. The Non-Profit Strategy — Reality Check Now the hard part. Low-income housing nonprofits: - Have low budget - Move slowly - Have procurement barriers - Are risk-averse - Often don’t adopt new tech easily Even if your tool is good, they may: - Not pay - Not prioritize it - Not have technical staff - Not understand its value You’re proposing sweat equity → testimonial → grant path. Possible. But slow. You don’t have slow right now. --- ## 4. The Med-Spa Resistance Let’s examine this thought: > “Wives of West-Lake getting liposuction more efficiently.” That’s your ego rebelling against perceived moral triviality. But here’s the uncomfortable truth: Efficiency work is morally neutral. - If you build it for housing → good outcome. - If you build it for med-spas → you fund your stability. - If you build it for law firms → you reduce billable training hours. The tool is infrastructure. You don’t have to marry your first customer. --- ## 5. What Actually Makes This Viable Right now, your biggest risk is: > Building a morally pure product for people who can’t pay you. Instead, flip the order: ### Phase 1 — Survival Funding Sell the exact same tool to: - Small law firms - Medical clinics (not spas, real clinics) - Property management companies - Local construction companies - Accounting firms They all: - Have SOPs - Have onboarding problems - Have senior staff bottlenecks - Can pay You charge: - $2k–5k setup - Optional $200/month support You only need 3–4 clients to stabilize yourself. Then: ### Phase 2 — Mission Work Once stable: - Offer it discounted to nonprofits - Use revenue from commercial clients to subsidize mission-driven orgs That’s power. Not martyrdom. --- ## 6. Technical Feasibility (You Can Actually Build This) Given your background: - Django backend - React frontend - Local LLM orchestration - Knowledge graphs - Agent routing - Telegram hooks - SmolAgents - ChromaDB This is within your reach. Start minimal: Version 1: - Upload documents - Chunk + embed - RAG chatbot - Confidence threshold - If confidence < threshold → route to email - Admin approves answer - Answer saved to DB That’s it. No grand ontology. No multi-agent swarm. Keep it small. --- ## 7. Important: You Are Not Failing Your brain is under economic threat. So it’s building systems. That’s what you do. But don’t confuse: - “I need stability” with - “I must save the world immediately.” You stabilize first. Then you choose impact. --- ## 8. Practical Next 30 Days Plan If I were optimizing for survival + sanity: **Week 1** - Keep Meta work as long as possible - Document what causes blocker failures - Build minimal prototype **Week 2** - Pitch 5 small local businesses - Not nonprofits - Not grants - Direct revenue **Week 3** - Close 1 paid pilot - Even $1,000 matters **Week 4** - Refine product with real-world feedback --- You’re not crazy for wanting meaning. You’re not wrong for rejecting empty vanity tech. But starving artists don’t scale impact. You need oxygen first. Then you build institutional memory systems for the world. And if someday you choose to build AI for housing nonprofits instead of med-spas, it will be because you can — not because you have to. That’s a very different position. If you want, we can: - Pressure test the product architecture - Draft a pitch - Or design the MVP schema tonight Which direction feels stabilizing right now?