An AI agent can be thought of as a simple While-loop. It uses an LLM to select an action, executes that action, evaluates the result, and repeats the process until the task is complete. Let’s take a closer look at each of these components: Brain: The LLM is the core. It reads the situation, thinks, and decides what to do next. The big shift from chatbot to agent: the model isn't writing text anymore, it's making choices. Planning: Hard tasks need more than one step. Agents break them down using methods like Chain of Thought (think step by step), Tree of Thoughts (try options, pick the best), or Reflexion (learn from mistakes and retry). Planning turns a fuzzy goal into clear actions. Tools: An LLM without tools is a brain in a jar. Tools are functions the model can call, like web search, code execution, APIs, files, or browsers (often using the MCP standard). The model requests a tool, the system runs it, and the result comes back. Memory: Without memory, every turn starts from zero. Short-term memory is the context window. Long-term memory lives in vector stores, files, and knowledge bases. When the window fills up, agents summarize old turns and carry the summary forward. Loop: All four pieces work together in a cycle. The agent looks at the current state, decides what to do, uses a tool, sees the result, and repeats. It keeps going until it gives a final answer. Guardrails: Not strictly anatomy, but important. Sandboxing, human checks, token limits, output validation, and scope limits keep autonomy from turning into expensive chaos. The more autonomy you give, the more these matter. Over to you: when you build an agent, which of these five takes the most work to get right?

the three-tier memory of Hermes agent. AI agents forgets everything when your session ends. Hermes doesn't. it has three memory layers, each at a different speed. 𝘁𝗶𝗲𝗿 𝟭: 𝘁𝘄𝗼 𝘁𝗶𝗻𝘆 𝗺𝗮𝗿𝗸𝗱𝗼𝘄𝗻 𝗳𝗶𝗹𝗲𝘀 MEMORY.md (2,200 chars) and USER.md (1,375 chars). injected into the system prompt at session start as a frozen snapshot. MEMORY.md holds project conventions, tool quirks, lessons learned. USER.md holds your profile: name, communication style, skill level. these files are tiny on purpose. when MEMORY.md hits ~80% capacity, the agent consolidates: merges related entries, drops redundancy, keeps only the densest facts. natural selection pressure applied to memory. the files stay small, but what's inside gets sharper over time. 𝘁𝗶𝗲𝗿 𝟮: 𝗳𝘂𝗹𝗹-𝘁𝗲𝘅𝘁 𝘀𝗲𝘀𝘀𝗶𝗼𝗻 𝘀𝗲𝗮𝗿𝗰𝗵 (𝘀𝗾𝗹𝗶𝘁𝗲 + 𝗳𝘁𝘀𝟱) every conversation gets stored in SQLite with FTS5 indexing. the agent can search weeks of past sessions on demand. when the agent calls session_search: FTS5 ranks matches in ~10ms over 10,000+ docs, an LLM summarizes the top hits, and a concise result returns to context. tier 1 is always present but tiny. tier 2 has unlimited capacity but requires an active search. critical facts live in memory, everything else is searchable. 𝘁𝗶𝗲𝗿 𝟯: 𝗲𝘅𝘁𝗲𝗿𝗻𝗮𝗹 𝗺𝗲𝗺𝗼𝗿𝘆 𝗽𝗿𝗼𝘃𝗶𝗱𝗲𝗿𝘀 8 pluggable providers that run alongside tiers 1 and 2, never replacing them. three worth knowing: Honcho (dialectic user modeling, 12 identity layers), Holographic (local-first, HRR vectors, no external calls), and Supermemory (context fencing that prevents the same fact from being re-stored infinitely). when active, hermes auto-syncs every turn: prefetch before, sync after, extract at session end. 𝗵𝗼𝘄 𝘁𝗵𝗲𝘆 𝗰𝗼𝗺𝗽𝗼𝘀𝗲 𝗶𝗻 𝗮 𝘀𝗶𝗻𝗴𝗹𝗲 𝘁𝘂𝗿𝗻 this is the part most people miss. the tiers compose on every turn through a five-step cycle: 1. turn opens. tier 1 is already in prompt, tier 3 prefetches and prepends. 2. agent responds using all three tiers as context. 3. periodic nudge fires (~every 300s). the agent reflects: "has anything worth persisting happened?" if yes, it writes. if no, it returns silently. 4. memory written to MEMORY.md on disk. invisible this session because the prefix cache stays warm. 5. session closes. tier 2 logs the transcript, tier 3 extracts semantics. next session opens with the new state. agent memory today is either always-on but shallow (stuff everything in the prompt) or deep but passive (vector store that never fires at the right time). hermes composes across both: tiny always-present files for critical facts, full-text search for deep recall, external providers for semantic modeling, all orchestrated by a nudge that decides autonomously what's worth saving. the agent doesn't just store memories. it curates them under pressure. i wrote a full deep dive (article below) covering hermes agent's memory system, self-evolving skills, GEPA optimization, and how to set up multiple specialized agents on your machine.

Rebuilding the Rails: New Financial Infra @26Broadway 🏦⚡️ The financial stack is being rebuilt. Join us at the heart of NYC to deconstruct the future of Stablecoins, Payments, and On-chain Assets. 🤵 Hosted by ME Group & @soluluUSA 🤝 Co-hosted by @CoinfoundGroup, @finchip_ai, @bitpushnews, @SoulByte_HD, @POSX_Official, @Bitfi_Org, @fmgroupxyz 📅 May 11 | 5:30 PM – 10:00 PM 📍 Station 3, 26 Broadway 3rd Floor, NYC 🔗 RSVP: (Invite-only) Speakers ✨ 🎤 Opening Remarks: Jessica — CEO of ME Group 👥 Panel | Building the Next Financial Rails: Shaun Imran — Co-Founder & CMO, SOLULU Steven — Founder, Future Money Group Han — Founder, BitFi Vera — Founder, SoulByte Carter — Founder, POSX & Omninal

AI is not going to replace human beings,” said @mcannonbrookes speaking to @l2k on @wandb’s podcast. And it definitely stayed with me. AI & agents are only going to enable us to do more, build more, and help humanity realize its full potential. #TLVCPartner#

A tricky LLM interview question: Your RAG system scores 90% retrieval accuracy on 5k company docs. But scaling to 500k docs drops the accuracy to just 50%, with the same embedding model and retriever. Why did this happen? The simplest answer is that more documents mean more competition for the top-k retrieval slots. That is true, but it doesn't explain why accuracy drops this dramatically. The answer comes down to how enterprise docs are distributed in the embedding space. Today, a single product decision in a company generates meeting transcripts, Slack threads, Confluence docs, Jira tickets, and email threads. They are related to the same event, so they all land in a similar region of the embedding space. As the company operates over months, this pattern repeats for every project/customer/roadmap, and the embedding space fills up with clusters of closely related documents. But all related docs don't contain the same facts. → Slack thread covers the decision made → Jira has the implementation deadline → Confluence has the technical spec → Email thread has the customer request When a query is about a specific fact (like a deadline), the answer lives in one of those docs. At a 5K corpus size, there might be 3-5 docs touching that topic, and the correct one easily lands in the top-k results. But at a 500K corpus size, there could be 40-60 total docs, and the one containing the actual answer can easily get pushed out of the top-k by other topically relevant docs, degrading retrieval. A recent research paper from Onyx documented this. The researchers used their newly open-sourced EnterpriseRAG-Bench dataset. It has 500k+ synthetic enterprise documents spread across Slack, Gmail, Jira, GitHub, Confluence, Google Drive, HubSpot, Fireflies, and Linear, with realistic noise like misfiled documents, near-duplicates, and conflicting versions. They ran the same retrievers at five corpus sizes from 5K to 500K. → Vector search accuracy dropped from 90.7% at 5K documents to 50.6% at 500K docs. → BM25 degraded more gracefully, from 85.8% to 68.4%. → At every scale, higher neighborhood density in the embedding space monotonically correlated with lower recall. The practical implication here is that retrieval accuracy on a 5k test set tells you almost nothing about production-scale performance. Always test at a realistic volume to measure the neighborhood density in your embedding space to estimate how much headroom the retriever actually has. The entire EnterpriseRAG-Bench dataset (500K docs with questions, and the whole evaluation harness) is open-source. Run your retriever against it at 5K, then at 500K, and see where your own accuracy curve breaks. I have shared the GitHub repo in the replies.

Beta is officially open. We’re inviting early testers to explore a marketplace where AI skills can be launched, acquired, and used as on chain assets. The first 5 OG invite codes are now available. OG members receive: • 8,000 starting points • 2.0x points multiplier • Exclusive Senior invite code issuance privileges Reserved only for the earliest network contributors inside the ecosystem. Explore the beta: Join the official Telegram for beta access instructions, updates, and tester support: To enter: → Follow @FinChip_AI → Like + repost this post + Tag 3 friends → Comment: “I want a invite code” Entries close in 24 hours. 5 winners will be announced under this post and receive an exclusive OG invite code for the beta.

New for financial services: ready-to-run Claude agent templates for building pitches, conducting valuation reviews, closing the books at month-end, and more. Install them as plugins in Cowork and Claude Code, or use our cookbooks to run them in production as Managed Agents.

The Beta Giveaway begins now 🚀 Before the beta test officially starts, we’re running a giveaway for our early community. 20 winners will receive 20U each. To enter: 1. Like this post and quote repost with: “ Beta is coming” + tag 3 friends 2. Follow @finchip_ai 3. Join our Telegram and share a screenshot of your quote repost in the group The giveaway closes in 48 hours. Winners will be announced on X. Stay tuned for beta access details.

Palo Alto: Agent Skills are the New Apps 🤖⚡️ We cut through the noise in Palo Alto to answer one question: How do Agents actually prove their value? 🛠️ 3 Hard Truths from the Floor: Skills > Apps📱: Standardized "Skill Packages" are the new unit of AI value. The era of the chatbot is over; the era of Executable Agents is here. The Economy of Agents💸: Intelligence is nothing without Settlement. Standardizing Agent-to-Agent payments is the final piece of the puzzle. Web3 Infra is Back🛠️: From hardware to programmable layers, Web3 builders are providing the "skeleton" for the Agentic future. Less talk, more execution. See you at the next one! 🚀 Special thanks to our co-hosts: @finchip_ai @MetaEraHK @iPolloClaw @GPTDAOCN @op_catlayer @OnePieceLabs #AgenticAI# #SiliconValley# #PaloAlto# #Web4# #AIAgents# #AgenticEconomy#

ANTHROPIC JUST RELEASED THE OFFICIAL PLAYBOOK FOR BUILDING A COMPANY WITH CLAUDE CODE. CEO: 1 human. Employees: AI agents. Operations: fully automatic. The zero-headcount company is no longer a joke.

Why credit cards can't power the agent economy, and what will. At the @USC VanEck Southern California Blockchain Conference, our Co-founder & CEO @ChiZhangData broke down why existing payment infrastructure fails when AI agents transact: ▷ Privacy risk: Agents need your CVV, name, and card number with zero accountability. ▷ Fraud triggers: Automated browser clicks get flagged by Visa/Chase algorithms instantly. ▷ No agent identity: Agents can't sign up or sign in. Humans must intervene every time. The fix? Stablecoin-native rails with smart contract delegation. Giving agents a specific amount, a specific purpose, and a specific time window. Not your whole wallet. This is what Kite is building.