Long-form on agent SDK design, typed memory, Plan Mode, and on-prem deployment.https://onto.neullabs.com/enhttps://onto.neullabs.com/blog/why-agent-memory-is-broken/https://onto.neullabs.com/blog/why-agent-memory-is-broken/The standard agent loop has no idea who you are between sessions. Conversation buffers forget, embeddings retrieve the similar but not the true. Here is what typed memory looks like instead.Mon, 25 May 2026 00:00:00 GMTagent memoryarchitecturethesishttps://onto.neullabs.com/blog/plan-mode-explained/https://onto.neullabs.com/blog/plan-mode-explained/How a first-class Plan Mode in your agent SDK turns hallucinated side effects into reviewable tasks — and why this is different from human-in-the-loop bolted on after the fact.Fri, 22 May 2026 00:00:00 GMTPlan Modehuman-in-the-loopagent safetyhttps://onto.neullabs.com/blog/typed-memory-vs-embeddings/https://onto.neullabs.com/blog/typed-memory-vs-embeddings/Why storing 'Alice is in UTC' as an embedding and retrieving it later by cosine similarity is the wrong primitive — and what to use instead.Wed, 20 May 2026 00:00:00 GMTtyped memoryvector searchagent memoryhttps://onto.neullabs.com/blog/open-weight-defaults/https://onto.neullabs.com/blog/open-weight-defaults/Why an agent SDK that defaults to a proprietary hosted model — and treats open-weight as the fallback — pushes its users into vendor lock-in and gives them no on-prem path.Mon, 18 May 2026 00:00:00 GMTopen-weightOllamamodel providershttps://onto.neullabs.com/blog/stateless-agents-stateful-humans/https://onto.neullabs.com/blog/stateless-agents-stateful-humans/The autonomous-loop architecture conflates two responsibilities: deciding and doing. Splitting them — agent stateless, human stateful — fixes more than it complicates.Fri, 15 May 2026 00:00:00 GMTagent architecturethesisdesignhttps://onto.neullabs.com/blog/agent-sdk-comparison-2026/https://onto.neullabs.com/blog/agent-sdk-comparison-2026/An honest head-to-head of the four major agent SDKs — where each wins, where each loses, and how to pick the right one for your project.Tue, 12 May 2026 00:00:00 GMTcomparisonClaude SDKOpenAI SDKGoogle ADKhttps://onto.neullabs.com/blog/onto-vs-langchain/https://onto.neullabs.com/blog/onto-vs-langchain/LangChain is the largest agent orchestration ecosystem in the world. So why pick ONTO instead? A direct comparison on memory, planning, consent, and operational concerns.Sun, 10 May 2026 00:00:00 GMTcomparisonLangChainagent memoryhttps://onto.neullabs.com/blog/onto-vs-llamaindex/https://onto.neullabs.com/blog/onto-vs-llamaindex/LlamaIndex is the most popular RAG framework. ONTO is an agent SDK. They solve adjacent problems — but treating them as substitutes leads to bad architecture. Here's how to pick.Fri, 08 May 2026 00:00:00 GMTcomparisonLlamaIndexRAGhttps://onto.neullabs.com/blog/onto-vs-mem0/https://onto.neullabs.com/blog/onto-vs-mem0/Mem0 is a popular dedicated memory layer for LLMs. ONTO is an agent SDK with memory built in. What's the difference, and which one is right for your project?Wed, 06 May 2026 00:00:00 GMTcomparisonMem0agent memoryhttps://onto.neullabs.com/blog/onto-vs-letta/https://onto.neullabs.com/blog/onto-vs-letta/Letta (formerly MemGPT) pioneered the idea of agents managing their own memory through tool calls. ONTO takes a typed-graph approach. Same problem, different solutions.Mon, 04 May 2026 00:00:00 GMTcomparisonLettaMemGPTagent memoryhttps://onto.neullabs.com/blog/onto-vs-crewai/https://onto.neullabs.com/blog/onto-vs-crewai/CrewAI made multi-agent orchestration accessible. ONTO takes a different shape — subagents under explicit consent scopes, a depth budget, and rolled-up cost. Which fits your problem?Sat, 02 May 2026 00:00:00 GMTcomparisonCrewAImulti-agenthttps://onto.neullabs.com/blog/onto-vs-autogen/https://onto.neullabs.com/blog/onto-vs-autogen/AutoGen pioneered conversational multi-agent and human-in-the-loop patterns. ONTO takes a different approach to the same problem — Plan Mode as a runtime invariant. Here's how they compare.Thu, 30 Apr 2026 00:00:00 GMTcomparisonAutoGenhuman-in-the-loophttps://onto.neullabs.com/blog/hipaa-ready-ai-agents/https://onto.neullabs.com/blog/hipaa-ready-ai-agents/A practical checklist for taking an AI agent into a HIPAA-covered environment — what to log, what to tag, what to gate, and what to deploy on-prem.Tue, 28 Apr 2026 00:00:00 GMThealthcareHIPAAcompliancehttps://onto.neullabs.com/blog/financial-services-ai-agents/https://onto.neullabs.com/blog/financial-services-ai-agents/Financial services AI deployment is harder than healthcare in some ways and easier in others. A practical guide to the SOC 2, FINRA, and GDPR controls your agent system has to satisfy.Sun, 26 Apr 2026 00:00:00 GMTfinancial servicesFINRASOC 2compliancehttps://onto.neullabs.com/blog/legal-ai-agents/https://onto.neullabs.com/blog/legal-ai-agents/Privilege, confidentiality, conflicts checking, and the unauthorized practice of law — what your legal AI agent actually has to handle, and how to handle it cleanly.Fri, 24 Apr 2026 00:00:00 GMTlegalconfidentialitycompliancehttps://onto.neullabs.com/blog/customer-support-no-hallucinated-refunds/https://onto.neullabs.com/blog/customer-support-no-hallucinated-refunds/How to ship a customer support agent that remembers each customer, proposes refunds and escalations through Plan Mode, and auto-approves only when policy says so.Wed, 22 Apr 2026 00:00:00 GMTcustomer supportmulti-tenantPlan Modehttps://onto.neullabs.com/blog/personal-ai-that-remembers/https://onto.neullabs.com/blog/personal-ai-that-remembers/Long-running personal AI fails on memory more than anything else. Here's an architecture for an assistant that learns who you are across sessions, with the user's facts staying on the user's side.Mon, 20 Apr 2026 00:00:00 GMTpersonal AIagent memoryasync writeshttps://onto.neullabs.com/blog/research-agents-typed-claims/https://onto.neullabs.com/blog/research-agents-typed-claims/A research copilot that pulls top-k chunks is fine. One that extracts typed claims with provenance, detects cross-paper conflicts, and asks before resolving them is better.Sat, 18 Apr 2026 00:00:00 GMTresearchknowledge graphextractionhttps://onto.neullabs.com/blog/hr-ai-gdpr/https://onto.neullabs.com/blog/hr-ai-gdpr/AI in recruiting and HR has specific failure modes — bias, automated decisions on candidates, retention of personal data. Here's how to ship an HR agent that doesn't trip Article 22, 5, or 6.Thu, 16 Apr 2026 00:00:00 GMTHRrecruitingGDPRcompliancehttps://onto.neullabs.com/blog/on-premise-ai-agents/https://onto.neullabs.com/blog/on-premise-ai-agents/A practical guide to running production AI agents fully on your infrastructure — local models, local storage, local embeddings, Helm charts. No third-party API calls.Tue, 14 Apr 2026 00:00:00 GMTon-premOllamaself-hostedopen-weighthttps://onto.neullabs.com/blog/multi-tenant-agent-memory/https://onto.neullabs.com/blog/multi-tenant-agent-memory/If you're running an agent SDK in a multi-tenant SaaS, memory leakage between tenants is the bug that ends your company. Here's the architecture that prevents it structurally.Sun, 12 Apr 2026 00:00:00 GMTmulti-tenantSaaSagent memoryhttps://onto.neullabs.com/blog/cost-observability-ai-agents/https://onto.neullabs.com/blog/cost-observability-ai-agents/The OpenAI bill arrives. You can't tell which customer or which workflow drove it. Here's how to instrument AI agents so cost attribution is built in — not reverse-engineered from logs.Fri, 10 Apr 2026 00:00:00 GMTcost observabilityFinOpsmulti-tenanthttps://onto.neullabs.com/blog/ontology-extraction-four-levels/https://onto.neullabs.com/blog/ontology-extraction-four-levels/Pulling typed facts from natural language is half the problem. The other half is knowing which level the fact lives at. Here's the four-level model and why every agent SDK should have one.Wed, 08 Apr 2026 00:00:00 GMTontology extractiontyped memoryexplainerhttps://onto.neullabs.com/blog/python-quickstart/https://onto.neullabs.com/blog/python-quickstart/Install ONTO, run your first agent, extract typed facts, plan an action, and execute it under consent scopes. No API key required.Mon, 06 Apr 2026 00:00:00 GMTtutorialPythongetting started