Free, open-source AI education; new modules as they are published.https://ai.hmohamedansari.com/en-gbhttps://ai.hmohamedansari.com/learn/automation-to-agents/automation-is-not-a-dirty-word/https://ai.hmohamedansari.com/learn/automation-to-agents/automation-is-not-a-dirty-word/Before we talk about models or agents, we need to remember what dependable automation already looks like.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/automation-to-agents/evidence-before-explanation/https://ai.hmohamedansari.com/learn/automation-to-agents/evidence-before-explanation/A good incident explanation is built from observable facts, not from the most confident-sounding story.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/automation-to-agents/rules-decisions-and-state/https://ai.hmohamedansari.com/learn/automation-to-agents/rules-decisions-and-state/A rule chooses an outcome. State tells the system where it is in a longer piece of work.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/automation-to-agents/events-queues-and-workflows/https://ai.hmohamedansari.com/learn/automation-to-agents/events-queues-and-workflows/Not every task happens in one request. Learn the machinery that lets work continue safely across time and failure.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/automation-to-agents/reliability-is-behaviour/https://ai.hmohamedansari.com/learn/automation-to-agents/reliability-is-behaviour/A dependable system has a clear answer for success, failure, delay, duplication and uncertainty.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/automation-to-agents/build-the-deterministic-investigator/https://ai.hmohamedansari.com/learn/automation-to-agents/build-the-deterministic-investigator/Put the pieces together: evidence, rules, state and safe outcomes in the first useful Workbench milestone.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/automation-to-agents/where-the-model-belongs/https://ai.hmohamedansari.com/learn/automation-to-agents/where-the-model-belongs/An LLM is a useful but uncertain component. The system around it still owns the decision, the tools and the consequences.Thu, 03 Sep 2026 00:00:00 GMTautomation-to-agentsstablehttps://ai.hmohamedansari.com/learn/advanced/a-harness-becomes-an-agent/https://ai.hmohamedansari.com/learn/advanced/a-harness-becomes-an-agent/An agent is not a chat window with tools. It is a bounded control loop that can observe, propose, act and stop under rules that somebody can inspect.Thu, 03 Sep 2026 00:00:00 GMTadvancedemerginghttps://ai.hmohamedansari.com/learn/advanced/context-is-a-budget-not-a-brain/https://ai.hmohamedansari.com/learn/advanced/context-is-a-budget-not-a-brain/A model can only work from the information supplied in this turn. Good context is selected evidence with a job to do, not every document you can fit into a window.Thu, 03 Sep 2026 00:00:00 GMTadvancedemerginghttps://ai.hmohamedansari.com/learn/advanced/memory-working-durable-retrieved/https://ai.hmohamedansari.com/learn/advanced/memory-working-durable-retrieved/Memory is not one magical feature. Separate what is needed now, what the system must retain as a record, and what it may retrieve later.Thu, 03 Sep 2026 00:00:00 GMTadvancedemerginghttps://ai.hmohamedansari.com/learn/advanced/tool-calls-are-proposals/https://ai.hmohamedansari.com/learn/advanced/tool-calls-are-proposals/A tool call from a model is structured input to your application. Validate it, authorize it and record it before it changes anything.Thu, 03 Sep 2026 00:00:00 GMTadvancedemerginghttps://ai.hmohamedansari.com/learn/advanced/mcp-without-the-fog/https://ai.hmohamedansari.com/learn/advanced/mcp-without-the-fog/MCP is a way for an AI application to discover and use declared capabilities. It does not replace your authorization, your API design or your production controls.Thu, 03 Sep 2026 00:00:00 GMTadvancedvolatilehttps://ai.hmohamedansari.com/learn/advanced/a2a-handing-work-to-another-system/https://ai.hmohamedansari.com/learn/advanced/a2a-handing-work-to-another-system/A2A is about one agentic system asking another to perform a bounded task. It is different from giving a host direct access to a tool.Thu, 03 Sep 2026 00:00:00 GMTadvancedvolatilehttps://ai.hmohamedansari.com/learn/advanced/multi-agent-systems-roles-delegation-evidence/https://ai.hmohamedansari.com/learn/advanced/multi-agent-systems-roles-delegation-evidence/Specialized roles can help when they create a useful boundary. They create theatre when they merely split one prompt into several names.Thu, 03 Sep 2026 00:00:00 GMTadvancedemerginghttps://ai.hmohamedansari.com/learn/advanced/gateways-policy-and-an-understandable-system/https://ai.hmohamedansari.com/learn/advanced/gateways-policy-and-an-understandable-system/A gateway can make cross-cutting controls consistent, but it cannot decide your architecture for you. Put identity, policy, limits and evidence at boundaries you can operate.Thu, 03 Sep 2026 00:00:00 GMTadvancedvolatilehttps://ai.hmohamedansari.com/learn/agents-in-production/production-is-a-different-job/https://ai.hmohamedansari.com/learn/agents-in-production/production-is-a-different-job/A useful prototype proves an idea. A production service makes an ongoing promise to people who depend on it, including a promise about failure.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionstablehttps://ai.hmohamedansari.com/learn/agents-in-production/golden-signals-for-agent-systems/https://ai.hmohamedansari.com/learn/agents-in-production/golden-signals-for-agent-systems/Start with latency, traffic, errors and saturation. Then add agent-specific measures such as quality and cost without losing the SRE signals that explain the service.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionstablehttps://ai.hmohamedansari.com/learn/agents-in-production/trace-the-whole-decision/https://ai.hmohamedansari.com/learn/agents-in-production/trace-the-whole-decision/A trace should let an operator follow one request through context selection, model calls, tools, policy decisions and final outcome without exposing secrets or private content.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionemerginghttps://ai.hmohamedansari.com/learn/agents-in-production/kubernetes-is-an-execution-environment/https://ai.hmohamedansari.com/learn/agents-in-production/kubernetes-is-an-execution-environment/Kubernetes can schedule and recover workloads. It cannot choose your task boundaries, retries, data controls or release policy for you.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionstablehttps://ai.hmohamedansari.com/learn/agents-in-production/grafana-dashboards-that-lead-to-a-decision/https://ai.hmohamedansari.com/learn/agents-in-production/grafana-dashboards-that-lead-to-a-decision/A useful dashboard answers an operational question quickly. It does not try to display every metric your service has ever emitted.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionstablehttps://ai.hmohamedansari.com/learn/agents-in-production/opamp-operating-the-telemetry-fleet/https://ai.hmohamedansari.com/learn/agents-in-production/opamp-operating-the-telemetry-fleet/Open Agent Management Protocol helps manage OpenTelemetry Collector-style agents at scale. It is about operating telemetry components, not about an AI agent making decisions.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionemerginghttps://ai.hmohamedansari.com/learn/agents-in-production/reliability-control-loops/https://ai.hmohamedansari.com/learn/agents-in-production/reliability-control-loops/Timeouts, retries, queues, idempotency and human escalation work together. A retry without a stop condition is just a faster way to make a bad incident bigger.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionstablehttps://ai.hmohamedansari.com/learn/agents-in-production/security-boundaries-for-agentic-systems/https://ai.hmohamedansari.com/learn/agents-in-production/security-boundaries-for-agentic-systems/Treat model input as untrusted, tool authority as sensitive and every integration as a supply-chain decision. Security is enforced by systems, not by asking a model to behave.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionemerginghttps://ai.hmohamedansari.com/learn/agents-in-production/release-and-incident-readiness/https://ai.hmohamedansari.com/learn/agents-in-production/release-and-incident-readiness/Before a model-enabled feature is trusted in production, prove that it can be evaluated, observed, constrained, rolled back and supported when it fails.Thu, 03 Sep 2026 00:00:00 GMTagents-in-productionemerging