MetisMesh — BlogsField notes from building real systems: agentic architecture, retrieval, governance, and how language models work.https://www.metismesh.comen-usArchitecting an Agentic Claims Assistant: A Simple, Grounded Explanation of Autonomous Insurance Triagehttps://www.metismesh.com/blogs/agentic_claims_assistanthttps://www.metismesh.com/blogs/agentic_claims_assistantA step-by-step, plain-language walkthrough of how to design a stateful, tool-using AI claims assistant using LangGraph — covering orchestration, shared state, multi-step reasoning, human-in-the-loop controls, and structured resolution outputs.Fri, 27 Feb 2026 00:00:00 GMTInsurance AI ArchitectureAgentic AIInsurance TechnologyClaims AutomationLangGraphWorkflow OrchestrationMulti-Step ReasoningHuman-in-the-LoopAI Governancekhalid.rizvi@metismesh.com (Khalid Rizvi)Compliant AI Orchestration in Insurance: From API Calls to Structured Claim Decisionshttps://www.metismesh.com/blogs/insurance_domain_agentic_orchestrationhttps://www.metismesh.com/blogs/insurance_domain_agentic_orchestrationA practical guide to building a stateful, agent-driven insurance workflow that orchestrates eligibility checks, contract validation, shared state management, human oversight, and data-grounded claim summaries.Fri, 27 Feb 2026 00:00:00 GMTApplied Generative AIAgentic AIWorkflow OrchestrationInsurance TechnologyClaims AutomationLangGraphHuman-in-the-Loopkhalid.rizvi@metismesh.com (Khalid Rizvi)Observability and Governance in RAG: Building Trustworthy AI for Insurance and Bankinghttps://www.metismesh.com/blogs/insurance_governance_observabilityhttps://www.metismesh.com/blogs/insurance_governance_observabilityA practical guide to understanding observability, governance, RAG validation, judge models, and safety guardrails—explained in simple terms for building compliant AI systems in insurance, banking, and regulated industries.Fri, 27 Feb 2026 00:00:00 GMTResponsible AIResponsible AIRAGAI GovernanceObservabilityInsurance TechnologyBanking AIComplianceRisk Managementkhalid.rizvi@metismesh.com (Khalid Rizvi)Building True AI Agents: Understanding the Core Architecturehttps://www.metismesh.com/blogs/ai_agent_core_architecture_overviewhttps://www.metismesh.com/blogs/ai_agent_core_architecture_overviewA clear, layer-by-layer explanation of how modern AI agents are constructed — from probabilistic token prediction to structured tool use, strong typing, and observable orchestration loops.Wed, 04 Feb 2026 00:00:00 GMTAI AgentsAI AgentsAgentic AILLMsPrompt EngineeringTool UseFunction CallingStructured OutputsPydantickhalid.rizvi@metismesh.com (Khalid Rizvi)Understanding and Implementing Simple Linear Regression from Scratchhttps://www.metismesh.com/blogs/linear_regressionhttps://www.metismesh.com/blogs/linear_regressionA grounded, intuitive introduction to simple linear regression using real-world farming analogies and first-principles reasoning.Fri, 30 Jan 2026 00:00:00 GMTMachine Learning FundamentalsLinear RegressionMachine LearningStatisticsFundamentalskhalid.rizvi@metismesh.com (Khalid Rizvi)What Is an AI Agent? A Simple Guide for New Learnershttps://www.metismesh.com/blogs/what_is_ai_agenthttps://www.metismesh.com/blogs/what_is_ai_agentA clear, story-style introduction to AI agents, how they use Generative AI and large language models, and why they feel more like smart helpers than simple chatbots.Tue, 27 Jan 2026 00:00:00 GMTAI FundamentalsAI AgentsGenerative AILLMsBeginner Friendlykhalid.rizvi@metismesh.com (Khalid Rizvi)Level Up Your AI Chats: Six Prompt Moves That Wake Your Words Uphttps://www.metismesh.com/blogs/six_prompt_trickshttps://www.metismesh.com/blogs/six_prompt_tricksSix small prompt shifts turn AI from a dull answer machine into a sharp thinking partner you can use all day, from budgets to bedtime stories.Mon, 26 Jan 2026 00:00:00 GMTAI Writingpromptinggenerative-aiproductivitywritingkhalid.rizvi@metismesh.com (Khalid Rizvi)How LLMs Choose the Next Token: A Practical Guide to Logits, Decoding, and Samplinghttps://www.metismesh.com/blogs/understanding_logits_and_probabilitieshttps://www.metismesh.com/blogs/understanding_logits_and_probabilitiesAn intuition-first guide to how large language models turn raw logits into next-token choices, using real-world analogies to explain temperature, top-k, and top-p.Sun, 25 Jan 2026 00:00:00 GMTMachine LearningLLMLogitsDecodingSamplingTop-KTop-PTemperatureNLPMachine Learningkhalid.rizvi@metismesh.com (Khalid Rizvi)How LLMs Choose the Next Token: A Practical Guide to Decoding and Samplinghttps://www.metismesh.com/blogs/guide_llm_decoding_paramshttps://www.metismesh.com/blogs/guide_llm_decoding_paramsAn intuition-first explanation of how large language models select the next token, demystifying temperature, top-k, and top-p through real-world analogies.Fri, 16 Jan 2026 00:00:00 GMTMachine LearningLLMDecodingSamplingTop-KTop-PTemperatureNLPMachine Learningkhalid.rizvi@metismesh.com (Khalid Rizvi)The Lifecycle of an LLMhttps://www.metismesh.com/blogs/llm_training_generation_lifecyclehttps://www.metismesh.com/blogs/llm_training_generation_lifecycleFrom raw text to real-time answers: how large language models actually work.Mon, 12 Jan 2026 00:00:00 GMTArchitectureEventsReasoningSignalskhalid.rizvi@metismesh.com (Khalid Rizvi)RAG with Tabular Data: From Schemas to Answershttps://www.metismesh.com/blogs/rag_tabular_datahttps://www.metismesh.com/blogs/rag_tabular_dataHow to design and operate Retrieval-Augmented Generation systems over relational and financial tabular data.Fri, 02 Jan 2026 00:00:00 GMTArchitectureRAGTabular DataFintechSQLArchitecturekhalid.rizvi@metismesh.com (Khalid Rizvi)