Asteri
@Asteri_eth • 2,266 subscribers
AI writer | Prediction markets researcher | the best vibe and fine content creator, as well as shitposts 🌌
Shorts
Videos
Karpathy found a way to reduce token consumption by 90% The problem is that the LLM re-reads the same files over and over again, loses context between documents, and provides less accurate answers as a result The solution is called Wiki Layer the LLM cleans, structures, and links all your data once, after which it never works with raw files again Three folders `raw/` for originals, `wiki/` for a clean knowledge base in Markdown, and files with rules for the agent Result up to 90% token savings on repeat queries, automatic links between documents, and a visual knowledge graph in Obsidian Everything stays on your local machine nothing goes to the cloud
Asteri1,094,858 views • 2 months ago
NVIDIA will refund your cloud bill if you let them put a $250,000 supercomputer on your desk for $2,999 The $2,999 DGX Spark puts a 128GB AI supercomputer on your desk the same 70B models you rented in the cloud, but nothing goes over the network and no ToS governs the machine you own $22,000 in annual savings is the big number everyone posts The quiet number is even bigger it’s the contracts you’ve stopped losing due to data residency Owning your own compute is no longer just budget optimization it’s a way to sell to customers who can’t send their data to the cloud
Asteri143,538 views • 2 months ago
A DENTAL CLINIC CAN LOSE $20K BEFORE THE PATIENT EVEN CALLS Not because the doctor is bad Because the first thing people see is a pattern: "nobody answered", "pricing was unclear", "they rushed me", "front desk was rude", "I waited forever", "they never followed up" That is enough for a buyer to choose the clinic two blocks away The interesting part is that Claude can turn those complaints into a sales asset for the business owner It reads the reviews, finds the exact phrases that keep killing trust, shows where the booking process breaks, then gives the team new replies, intake scripts, FAQ pages, follow-up messages, and a weekly view of what changed This is a cleaner offer than "we will get you more leads" You are showing them the leads they are already losing For high-ticket local businesses, that is easier to understand than another ad campaign, because the proof is sitting on their Google profile in the customer’s own words Full article below
Asteri16,199 views • 1 month ago
MOST PEOPLE USE CLAUDE CODE LIKE CHATGPT WITH A TERMINAL That is why they only get 30% of the value. They paste a task, wait for an answer, then keep explaining the same rules again when the context gets messy. The real power starts when you stop treating it like a chatbot and start treating it like an operating system for work. The useful commands are not flashy. They just remove the parts that quietly waste hours: > /init builds the project memory > /memory edits the rules Claude should keep > /clear resets the chat without losing the setup > /compact compresses a long session before quality drops > /context shows where your tokens are going > /rewind rolls back when an edit breaks the direction > /plan forces a blueprint before execution > /model matches the brain to the task > /goal lets it work toward a clear finish line This is the difference between asking AI for help and actually managing an AI worker. One mode gives you random useful answers. The other gives you memory, boundaries, checkpoints, context control and repeatable execution. Most people are still trying to write better prompts. The better move is building a better system around the model. save this
Asteri15,241 views • 1 month ago
Traditional development agencies are walking dead, they just don't know it yet Typical agency burns 70% of revenue on salaries for work that's repetitive and replaceable Kimi K2.6 lets one person run a 300-agent swarm for $500/mo and scale to $80k/mo Not a chatbot a fully autonomous execution layer that solves 2/3 of real GitHub issues out of the box
Asteri17,547 views • 2 months ago
ONE AI AGENT CAN TURN EMPTY BACKYARDS INTO POOL CONTRACTOR LEADS BEFORE THE HOMEOWNER EVEN ASKS FOR A QUOTE It works like this: AI scans expensive homes, finds large backyards without pools, generates a realistic pool preview, estimates the potential value increase, and turns it into a personalized sales pitch. It’s a fully automated workflow that: > Finds homes worth $600K-$2M > Detects large unused backyards with no pool > Generates a realistic before/after pool visualization > Estimates project cost and potential home value upside > Turns everything into a ready-to-send offer for the homeowner or contractor The valuable part is not that AI can make a pretty image. It’s that the buyer sees the outcome before anyone sells it to them. That changes the whole offer. You are not saying "we build pools." You are showing the exact pool that could exist in their backyard. And in the article below you’ll find 4 more AI automations that can turn boring workflows into real revenue
Asteri10,900 views • 1 month ago
THIS OPERATOR REPLACED A $50,000 AUTOMATION AGENCY WITH A 3-HOUR BUIL 378 browser tabs. 47 Notion pages nobody touched in 6 months. A spreadsheet that crashes every Monday He didn't hire a dev team. He didn't buy n8n enterprise He cloned a GitHub repo, wired 3 API keys, and typed /onboard into Claude Code By day 14 his team messaged the AI OS in Slack instead of messaging him His old workflow: 3 hours of context switching, 2 hours of actual work His new workflow: 5 hours of deep work, 30 minutes of admin This is not a dashboard. This is a command center Full breakdown below ↓
Asteri10,194 views • 2 months ago
No more content to load