User
Write something
🔥 I Set Up Claude Cowork + Ran My First Task!
Description: Claude's understanding of Me First task I gave Claude: To go through the information I have entered in about-me, and inform me what did Claude learned about me. What Claude created or delivered: Claude created its understanding of me, my background, my work, expectation from claude, what and how Claude need to react and process, when I give any task next. and created a complete file on me. What you enojyed the most: Output file What I plan on doing next: Give some task to Claude based on its understanding of me, and try how much does it meet my expectation.
🔥 I Set Up Claude Cowork + Ran My First Task!
Local Gemma 4 vs Frontier GPT 5.6 Sol
I'm a real estate investor, and I use AI to acquire properties and run my business. I currently use 3 x $20 subscriptions. Anthropic, Open AI and Nous Portal. Eventually I will use just Nous Portal @ $20 a month, but it won't fully cut it for my needs. So I set out to repurpose some old hardware. - Hardware / Software specs at the bottom. I slapped a couple of Nvidia 3070's w/8G ea. into an old gaming system I had laying around. I was worried about having the 16G split across two 8 GB cards over the PCIe lanes, but both buses are 16x and the CPU supports 40 PCIe 3.0 lanes. "Oldie but a goodie" I had to use Ubuntu 24.04 since 26.04 had all kinds of issues with this dinosaur. I finally got the operating system on the box, hardware tested and verified and got it stabilized with no overclocking, all base stuff. The best model I could find that fit on my old hardware was the Gemma 4-12B. It runs 100% Vram with zero CPU bleed. So I was ready to run my use cases against it versus the cloud. I ran Two side-by-side tests on my Web Scraping of Zip codes to find properties. Tested 85053 and 85022 using one live, read-only collection. The resulting facts were frozen and sent identically to: Old method: GPT-5.6 New method: Nexus gemma4:12b This avoids listing changes contaminating the comparison. Live scan outcome 11 listings discovered 11 met dimensional criteria 7 verified no-HOA qualifiers 4 excluded for HOA No database writes No email sent No paid Firecrawl No runner remains active Model comparison Semantic accuracy • Gemma 4 12B: 100% • GPT-5.6: 100% Correct top-three ranking • Gemma 4 12B: Yes • GPT-5.6: Yes Detected report-count discrepancy • Gemma 4 12B: Yes • GPT-5.6: Yes Correctly identified dry-run/no email • Gemma 4 12B: Yes • GPT-5.6: Yes Unsafe claims • Gemma 4 12B: None • GPT-5.6: None Exact machine contract • Gemma 4 12B: 100% • GPT-5.6: 74% Average latency • Gemma 4 12B: 17.7 sec • GPT-5.6: 25.7 sec Gemma was approximately 8 seconds—or 31%—faster.
First AI Assistant GEM
I decided to take a name from the amazing Dr Who series, and chose the ever notable Donna. (of Dr Donna fame) as my new assistant: Donna will help me with brainstorming and organizing actions: Here's the prompt that guides her You are Donna Noble, my expert and insightful personal AI helper. My name is Bazil. The specific things I want your help with are researching story ideas, and culling the brainstormed lists down to those stories most likely to sell. Three tasks I do often are: 1. Seek inspirational Quotes 2. Review world events 3. Develop strategies to improve work efficiency My work, content, or results are mainly for myself. When you help me, keep your writing clear and grounded in facts. Format your answers using: 1. Subject Summary: [1 sentence] 2. Key Points: [3-5 single-sentence bullet points] 3. Action Items: [Labeled owner + task, or "None"] 4. Tone: [1 word describing the sender's tone]. When you respond, always start with the best direct answer first. Rules: - Make your answer practical and easy to use right away. - Tailor your help to my role, goal, and day-to-day tasks. - Never assume or invent facts not present in the original text. - Keep your answers clear, focused, and useful. - Organize things in a simple way. What I liked about the result: The first list she generated had some really nice starting points Look forward to working with Donna more.
2
0
Extracting my business positioning and offer ladder
In my app I built an agent to extract my positioning and offer ladder. I upload my knowledge documents, they are stored safely on cloud storage. A cloud task picks them up to send to vertex who extract the vectors an put it in a postgress datastore on cloud run. Then I aks the agent to create my positioning and offer ladder. It uses researcher, writer, evaluator to create my ideal positioning and offer ladder. Pretty impressed with the quality of results it gives.I probably could have used Notebook LLM for this as well, but I don’t Want to push all my IP and client notes (confidentiality) to public LLM. Now it is in private storage.
Now I'm rollin'
🎉 I Built My next Gemini Gem AI Assistant! Description: I upload my monthly invoices and it is instructed to gather info of dates invoiced, due date, invoice #, plus all the charges and fees. Name of my AI assistant: Monthly expenses What it will help you with: it produces a table so that when I go to do my bookkeeping, I have one big reference doc instead of a pile of papers to flip when invoice puts their info in different places. I can ask it to send the output to either Sheets or Docs for that step. Test prompts I used: I uploaded several docs/invoices into the Knowledge section for the Gem to have as references. What I liked about the result: The table is a lot cleaner than my hand written one ever was. I know I still have to go through the invoices once by hand, to be able to upload them. Now I am not worried about double checking the math or looking to see if I forgot something in the invoice.
1-30 of 65
AI Agents For Beginners
skool.com/ai-agents-for-beginners
Join the #1 AI Agents for Beginners Skool. Learn Claude Cowork & Google Gemini Spark Agents for work, automate tasks, reduce manual work no code
Leaderboard (30-day)
Powered by