AI-enabled cloud platforms, agentic infrastructure, MCP servers, and self-healing systems for safety-critical industries — written and built by Ajin Baby, AI Platform & Cloud Infrastructure Architect.https://cloudandsre.com/https://cloudandsre.com/blog/acid-vs-base-and-isolation-levels/https://cloudandsre.com/blog/acid-vs-base-and-isolation-levels/ACID promises correctness; BASE trades it for scale — and your default isolation level is weaker than you think. A refresher on what 'consistent' means.Sun, 21 Sep 2025 00:00:00 GMTFundamentalsDistributed Systemshttps://cloudandsre.com/blog/agent-authorization-is-an-audit-trail-problem/https://cloudandsre.com/blog/agent-authorization-is-an-audit-trail-problem/NIST, a Senate bill and the MCP roadmap converge on one requirement: prove which agent did what, for whom, under whose grant. Your traces do not.Tue, 25 Aug 2026 00:00:00 GMTAIPlatform EngineeringSecurityhttps://cloudandsre.com/blog/agent-first-browser-cost-curve/https://cloudandsre.com/blog/agent-first-browser-cost-curve/Cloudflare's Kitesurf is a browser built for agents, not people. The measured trade-off, what it can't do, and why cheap browsing rewrites your threat model.Thu, 03 Sep 2026 00:00:00 GMTAIPlatform EngineeringSecurityhttps://cloudandsre.com/blog/agent-observability-what-to-instrument/https://cloudandsre.com/blog/agent-observability-what-to-instrument/A request either worked or didn't. An agent run can succeed on every span and still be wrong, slow and expensive. What the GenAI conventions ask you to record.Tue, 22 Sep 2026 00:00:00 GMTFundamentalsAIObservabilityhttps://cloudandsre.com/blog/agent-sandbox-cold-start-economics/https://cloudandsre.com/blog/agent-sandbox-cold-start-economics/Cloudflare cut median time-to-interactive from 4.0s to 648ms. The number matters because cold start is why teams share sandboxes they shouldn't share.Wed, 30 Sep 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/agent-skill-supply-chain/https://cloudandsre.com/blog/agent-skill-supply-chain/Agent skills are dependencies with none of the controls. What the 2026 research found, why scanners miss it, and seven fixes for platform teams.Sun, 02 Aug 2026 00:00:00 GMTAISecurityPlatform Engineeringhttps://cloudandsre.com/blog/agent-sprawl-is-your-next-production-incident/https://cloudandsre.com/blog/agent-sprawl-is-your-next-production-incident/Teams shipping AI agents are recreating 2015's microservices sprawl with worse observability. The governance surface that contains it before it pages you.Fri, 05 Jun 2026 00:00:00 GMTAIPlatform EngineeringReliabilityhttps://cloudandsre.com/blog/agent-watchdog-on-separate-silicon/https://cloudandsre.com/blog/agent-watchdog-on-separate-silicon/The Open Agent Safety Platform puts enforcement on a DPU, outside the host the agent runs on. The instinct is decades old. The shape is the question.Tue, 29 Sep 2026 00:00:00 GMTAIPlatform EngineeringSecurityhttps://cloudandsre.com/blog/agent-workflows-as-code/https://cloudandsre.com/blog/agent-workflows-as-code/OpenAI is winding down Agent Builder and platform Evals. Production agent workflows need code, tests, traces, and reviewable contracts.Wed, 05 Aug 2026 00:00:00 GMTAIPlatform EngineeringReliabilityhttps://cloudandsre.com/blog/agentic-ai-patterns-decision-guide/https://cloudandsre.com/blog/agentic-ai-patterns-decision-guide/Six agentic AI patterns — ReAct, Plan-and-Execute, Critic, fan-out, HITL gate, Supervisor — with a decision flowchart for picking one before you build.Fri, 26 Jun 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/agentic-ai-patterns-maturity-model/https://cloudandsre.com/blog/agentic-ai-patterns-maturity-model/A five-level agentic AI maturity model, from manual to multi-agent mesh — with a self-assessment and where regulated industries should draw the line.Fri, 26 Jun 2026 00:00:00 GMTAIPlatform EngineeringStrategyhttps://cloudandsre.com/blog/agentic-ai-patterns-production-failures/https://cloudandsre.com/blog/agentic-ai-patterns-production-failures/Every agentic AI pattern looks clean in a demo. Where each one breaks in production, the signals you're hitting them, and mitigations that actually work.Fri, 26 Jun 2026 00:00:00 GMTAIPlatform EngineeringReliabilityhttps://cloudandsre.com/blog/agentic-resource-discovery-ard/https://cloudandsre.com/blog/agentic-resource-discovery-ard/Google, Microsoft, and Hugging Face shipped Agentic Resource Discovery. The real ai-catalog.json schema, a working catalog, and the gotchas that break it.Thu, 02 Jul 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/agents-need-identities-not-api-keys/https://cloudandsre.com/blog/agents-need-identities-not-api-keys/Every AI agent is a non-human identity — most run on shared, long-lived API keys no IAM review sees. Per-agent identity and your credential blast radius.Wed, 08 Jul 2026 00:00:00 GMTAISecuritySREhttps://cloudandsre.com/blog/ai-sre-category-buyers-guide/https://cloudandsre.com/blog/ai-sre-category-buyers-guide/Gartner's first Market Guide for AI SRE tooling projects 85% adoption by 2029. Six evaluation questions for cutting through the category hype.Fri, 17 Jul 2026 00:00:00 GMTAISREPlatform Engineeringhttps://cloudandsre.com/blog/alert-fatigue-let-ai-triage/https://cloudandsre.com/blog/alert-fatigue-let-ai-triage/How I built alert-explainer — an open-source service that turns every Prometheus alert into a plain-English brief in 1–4 seconds for under a cent.Wed, 29 Apr 2026 00:00:00 GMTAISREOpen Sourcehttps://cloudandsre.com/blog/amazon-leadership-principles-my-thoughts/https://cloudandsre.com/blog/amazon-leadership-principles-my-thoughts/What Amazon's Leadership Principles actually mean in hiring and evaluations — and how to prepare for the behavioral rounds of an Amazon interview.Sat, 29 May 2021 00:00:00 GMTAWSGeneralLeadershiphttps://cloudandsre.com/blog/autonomous-agent-breach-control-list/https://cloudandsre.com/blog/autonomous-agent-breach-control-list/Hugging Face's July 2026 postmortem is the clearest agent-intrusion writeup we have. The failures were IMDS, admission policy, and one shared credential.Fri, 21 Aug 2026 00:00:00 GMTSecurityPlatform EngineeringAIhttps://cloudandsre.com/blog/backups-rpo-rto-restore-testing/https://cloudandsre.com/blog/backups-rpo-rto-restore-testing/RPO and RTO are promises you make before an incident and discover during one. Run the drill, time the restore, and find out which of your assumptions is wrong.Sat, 03 Oct 2026 00:00:00 GMTFundamentalsReliabilityhttps://cloudandsre.com/blog/big-o-notation-in-the-age-of-vector-search/https://cloudandsre.com/blog/big-o-notation-in-the-age-of-vector-search/Big-O still decides whether your system survives real data. A refresher on complexity, and why it governs vector search, context windows, and outages.Sun, 15 Jan 2023 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/bloom-filters-explained/https://cloudandsre.com/blog/bloom-filters-explained/A Bloom filter answers 'have I seen this before?' in kilobytes, never returning a false negative. How it works, and why databases and caches rely on it.Thu, 14 Dec 2023 00:00:00 GMTFundamentalsData Engineeringhttps://cloudandsre.com/blog/bounded-autonomy-for-ai-sre-agents/https://cloudandsre.com/blog/bounded-autonomy-for-ai-sre-agents/SREs face 50+ alerts a day at 60% false positives while vendors promise autonomous resolution. The autonomy ladder: what an AI agent should never do alone.Fri, 12 Jun 2026 00:00:00 GMTAIReliabilityArchitecturehttps://cloudandsre.com/blog/btree-vs-lsm-tree-storage-engines/https://cloudandsre.com/blog/btree-vs-lsm-tree-storage-engines/Almost every database is a read-optimized B-tree or a write-optimized LSM-tree underneath. Knowing which explains its performance and failure modes.Tue, 18 Apr 2023 00:00:00 GMTFundamentalsStoragehttps://cloudandsre.com/blog/building-incident-scribe/https://cloudandsre.com/blog/building-incident-scribe/How I built an open-source tool that turns messy Slack incident threads into blameless, structured incident reports in under 30 seconds.Tue, 21 Apr 2026 00:00:00 GMTAISREOpen Sourcehttps://cloudandsre.com/blog/caching-and-eviction-policies/https://cloudandsre.com/blog/caching-and-eviction-policies/The eviction policy decides whether your cache works. LRU vs LFU vs FIFO, hit rates, invalidation — and how the same bets govern prompt and KV caches.Sun, 28 Jan 2024 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/cap-theorem-in-ai-native-distributed-systems/https://cloudandsre.com/blog/cap-theorem-in-ai-native-distributed-systems/CAP didn't get repealed when LLMs showed up. How the C/A/P trade-offs shift when the datastore is a vector index, context graph, or retrieval layer.Tue, 12 May 2026 00:00:00 GMTAIArchitectureDistributed Systemshttps://cloudandsre.com/blog/capacity-is-your-top-ai-failure-mode/https://cloudandsre.com/blog/capacity-is-your-top-ai-failure-mode/Datadog found ~5% of production AI requests fail, nearly 60% on capacity limits. Provider saturation is an SRE problem you already have the primitives for.Tue, 25 Aug 2026 00:00:00 GMTReliabilityAIPlatform Engineeringhttps://cloudandsre.com/blog/cascading-failure-backpressure-load-shedding/https://cloudandsre.com/blog/cascading-failure-backpressure-load-shedding/A cascading failure is a feedback loop, not a row of dominoes. Removing the trigger does not stop it — which is why the controls have to be in place beforehand.Sat, 22 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/chaos-engineering-for-mcp/https://cloudandsre.com/blog/chaos-engineering-for-mcp/LLM calls fail 1–5% of the time and agent tasks fan out into 10–20 tool calls. How to fault-inject your MCP layer with mcp-chaos before production does.Sun, 21 Jun 2026 00:00:00 GMTAIReliabilityMCPhttps://cloudandsre.com/blog/claude-code-local-models/https://cloudandsre.com/blog/claude-code-local-models/Claude Code can point at Ollama, vLLM, LM Studio, or any private endpoint. When that's the right call, exactly how to configure it, and what you give up.Fri, 26 Jun 2026 00:00:00 GMTAIDeveloper ToolsPlatform Engineeringhttps://cloudandsre.com/blog/clock-skew-and-time-in-distributed-systems/https://cloudandsre.com/blog/clock-skew-and-time-in-distributed-systems/Wall clocks jump backwards, NTP corrects under you, and two machines never agree. What to use instead, and where time is load-bearing anyway.Wed, 23 Sep 2026 00:00:00 GMTFundamentalsDistributed Systemshttps://cloudandsre.com/blog/compaction-is-a-reliability-event/https://cloudandsre.com/blog/compaction-is-a-reliability-event/New research on long-horizon agents finds context compression degrades execution, not knowledge. Treat every compaction as a state transition and instrument it.Tue, 18 Aug 2026 00:00:00 GMTAIReliabilityPlatform Engineeringhttps://cloudandsre.com/blog/compilers-interpreters-and-jit/https://cloudandsre.com/blog/compilers-interpreters-and-jit/Compiling ahead of time, interpreting, or JIT-compiling hot paths explains why Python is slow and PyTorch is fast — torch.compile is the same idea.Tue, 19 Nov 2024 00:00:00 GMTFundamentalsDevhttps://cloudandsre.com/blog/concurrency-vs-parallelism/https://cloudandsre.com/blog/concurrency-vs-parallelism/Concurrency is dealing with many things at once; parallelism is doing them at once. A refresher on the GIL, async vs threads, and scaling model calls.Sun, 17 Mar 2024 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/config-changes-still-cause-the-biggest-outages/https://cloudandsre.com/blog/config-changes-still-cause-the-biggest-outages/Fresh postmortems keep confirming it: config changes, not code, cause the largest incidents. Agentic ops multiplies the volume. Here's the defense.Fri, 17 Jul 2026 00:00:00 GMTSREReliabilityPlatform Engineeringhttps://cloudandsre.com/blog/consistent-hashing-explained/https://cloudandsre.com/blog/consistent-hashing-explained/Hash-mod-N reshuffles almost everything when a node leaves; consistent hashing moves a fraction. The ring, virtual nodes, and why your CDN depends on it.Sun, 29 Oct 2023 00:00:00 GMTFundamentalsDistributed Systemshttps://cloudandsre.com/blog/context-engineering-the-window-is-a-budget/https://cloudandsre.com/blog/context-engineering-the-window-is-a-budget/The context window is your agent's working memory, not a junk drawer. Four operations — write, select, compress, isolate — and a token budget you allocate.Sat, 06 Jun 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/deadlocks-locks-and-race-conditions/https://cloudandsre.com/blog/deadlocks-locks-and-race-conditions/Races and deadlocks only fail sometimes — that's what makes them hard. Locks, atomicity, the four Coffman conditions, and why hung agents are the same bug.Mon, 06 May 2024 00:00:00 GMTFundamentalsReliabilityhttps://cloudandsre.com/blog/default-reasoning-model-lock-in/https://cloudandsre.com/blog/default-reasoning-model-lock-in/Salesforce made Claude the default model across Agentforce and Slack. Where model dependency actually accrues, and the five exits worth keeping open.Tue, 01 Sep 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/error-budgets-for-autonomy/https://cloudandsre.com/blog/error-budgets-for-autonomy/SRE solved runaway release risk with error budgets. The same mechanism governs AI agents: authority per action class, demoted fast, promoted slow.Mon, 06 Jul 2026 00:00:00 GMTAIReliabilitySREhttps://cloudandsre.com/blog/eu-ai-act-delay-platform-teams/https://cloudandsre.com/blog/eu-ai-act-delay-platform-teams/The EU AI Act omnibus moved high-risk deadlines to 2027–28. The logging, oversight, and inventory work is still yours — and reliability needs it anyway.Wed, 15 Jul 2026 00:00:00 GMTAIPlatform EngineeringReliabilityhttps://cloudandsre.com/blog/evaluating-ai-sre-agents-2026/https://cloudandsre.com/blog/evaluating-ai-sre-agents-2026/Datadog, Dynatrace, New Relic, AWS — every incumbent now ships an AI SRE agent. A field guide for evaluating one before it touches production.Sat, 11 Jul 2026 00:00:00 GMTAISREReliabilityhttps://cloudandsre.com/blog/floating-point-and-numerical-precision/https://cloudandsre.com/blog/floating-point-and-numerical-precision/Floating-point errors aren't random. Why 0.1 + 0.2 ≠ 0.3, and how the same fundamentals drive the FP32 → BF16 → FP8 march behind affordable LLMs.Tue, 01 Oct 2024 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/garbage-collection-explained/https://cloudandsre.com/blog/garbage-collection-explained/GC frees you from manual memory management — and occasionally freezes your program. Reachability, generational GC, and why GC tuning is a p99 problem.Sun, 12 Jan 2025 00:00:00 GMTFundamentalsReliabilityhttps://cloudandsre.com/blog/gettings-started-with-ansible/https://cloudandsre.com/blog/gettings-started-with-ansible/Ansible automates configuration and orchestration of infrastructure as code. A getting-started guide: inventory, playbooks, and your first runs.Wed, 26 May 2021 00:00:00 GMTAnsibleGeneralhttps://cloudandsre.com/blog/google-sre-autonomous-control-plane/https://cloudandsre.com/blog/google-sre-autonomous-control-plane/Google SRE splits autonomous incident response into reasoning, actuation, and evaluation. The split is the lesson — here's what to copy at normal scale.Fri, 17 Jul 2026 00:00:00 GMTAISREReliabilityhttps://cloudandsre.com/blog/graph-traversal-bfs-dfs/https://cloudandsre.com/blog/graph-traversal-bfs-dfs/BFS or DFS — queue or stack — decides everything downstream. A refresher on graph traversal, the visited set, and why GraphRAG is just a walk.Sun, 15 Jun 2025 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/graphql-for-api-development/https://cloudandsre.com/blog/graphql-for-api-development/GraphQL is a query language for APIs, created at Facebook in 2012. What it is, how it differs from REST, and where it fits in API development.Sat, 19 Nov 2022 00:00:00 GMTAPIFundamentalshttps://cloudandsre.com/blog/harness-engineering-the-third-phase/https://cloudandsre.com/blog/harness-engineering-the-third-phase/Agent = Model + Harness, and in 2026 the harness is the bottleneck. What a production-grade SRE harness contains, with a ~40-line reference implementation.Fri, 29 May 2026 00:00:00 GMTAISREOpen Sourcehttps://cloudandsre.com/blog/hash-tables-the-data-structure-behind-everything/https://cloudandsre.com/blog/hash-tables-the-data-structure-behind-everything/The hash table sits under your cache, index, dedup, and vector store metadata. How it turns a key into O(1) access — and what keeps it fast.Sun, 05 Mar 2023 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/hello-world/https://cloudandsre.com/blog/hello-world/First post: why I finally set up this blog after five or six years of thinking about it.Mon, 15 Feb 2021 00:00:00 GMTGeneralhttps://cloudandsre.com/blog/how-cdns-work/https://cloudandsre.com/blog/how-cdns-work/A CDN is a globe-spanning cache in front of your origin. The wins are latency and offload; the hard parts are cache keys, TTLs, and knowing when to invalidate.Mon, 21 Sep 2026 00:00:00 GMTFundamentalsNetworkshttps://cloudandsre.com/blog/how-dns-resolution-works/https://cloudandsre.com/blog/how-dns-resolution-works/DNS caching decides how fast failovers propagate and how far a misconfig spreads. The recursive walk from root to authoritative, and why TTLs govern deploys.Sun, 23 Jul 2023 00:00:00 GMTFundamentalsNetworkshttps://cloudandsre.com/blog/how-tcp-ip-works/https://cloudandsre.com/blog/how-tcp-ip-works/Every API call and model request rides on TCP/IP. The three-way handshake, flow control, and why round-trips and TIME_WAIT shape your latency.Wed, 07 Jun 2023 00:00:00 GMTFundamentalsNetworkshttps://cloudandsre.com/blog/idempotency-keys-explained/https://cloudandsre.com/blog/idempotency-keys-explained/A timeout never tells you whether the work happened. Idempotency keys resolve that — if you get the concurrency and scope right.Thu, 10 Sep 2026 00:00:00 GMTFundamentalsPlatform EngineeringAIhttps://cloudandsre.com/blog/idp-control-plane-for-agents/https://cloudandsre.com/blog/idp-control-plane-for-agents/Agents are becoming first-class platform citizens. That is an IDP problem: identity, quotas, golden paths, and a portal agents consume as an API.Mon, 31 Aug 2026 00:00:00 GMTAIPlatform EngineeringDeveloper Experiencehttps://cloudandsre.com/blog/inbound-agents-personal-agent-protocol/https://cloudandsre.com/blog/inbound-agents-personal-agent-protocol/Sierra and Meta propose a front door for agents acting on a customer's behalf. The hard questions are the ones the spec has not answered yet.Wed, 07 Oct 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/inference-gateway-model-aware-routing/https://cloudandsre.com/blog/inference-gateway-model-aware-routing/The Gateway API Inference Extension is GA. Why a normal Kubernetes Service balances model servers badly, and how to tell if you need an InferencePool.Sun, 02 Aug 2026 00:00:00 GMTAIPlatform EngineeringKuberneteshttps://cloudandsre.com/blog/inference-unit-economics/https://cloudandsre.com/blog/inference-unit-economics/You can run a GPU at 100% utilization and waste most of it. The cost model that predicts your inference bill, and the five levers in payback order.Sun, 02 Aug 2026 00:00:00 GMTAIPlatform EngineeringFinOpshttps://cloudandsre.com/blog/infrastructure-as-code-state-and-drift/https://cloudandsre.com/blog/infrastructure-as-code-state-and-drift/IaC is a three-way comparison: config, state, reality. Every confusing Terraform outcome is two of them disagreeing — and the plan tells you which.Wed, 12 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/introduction-to-yaml/https://cloudandsre.com/blog/introduction-to-yaml/YAML Ain't Markup Language — a human-readable format for configuration. The syntax, the gotchas, and why it's everywhere in infrastructure.Mon, 10 May 2021 00:00:00 GMTGeneralhttps://cloudandsre.com/blog/kubernetes-ai-conformance-checklist/https://cloudandsre.com/blog/kubernetes-ai-conformance-checklist/CNCF's Kubernetes AI Conformance program is a portability standard. Read it as a self-audit instead and it becomes a genuinely useful platform checklist.Thu, 20 Aug 2026 00:00:00 GMTPlatform EngineeringAIKuberneteshttps://cloudandsre.com/blog/load-balancing-and-health-checks/https://cloudandsre.com/blog/load-balancing-and-health-checks/Balancing decides where load goes; health checks decide where it stops. Most outages here come from a check that can't tell busy apart from broken.Fri, 11 Sep 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/mcp-2026-07-28-spec-what-changed/https://cloudandsre.com/blog/mcp-2026-07-28-spec-what-changed/The final MCP spec deletes sessions, mirrors requests into HTTP headers, and turns elicitation into multi-round-trip calls. What breaks, and what to audit.Wed, 29 Jul 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/mcp-a2a-two-layer-protocol-stack/https://cloudandsre.com/blog/mcp-a2a-two-layer-protocol-stack/A2A hit v1.0 with 150+ orgs behind it. MCP connects agents to tools; A2A connects agents to each other. What the two-layer stack means for platform teams.Mon, 13 Jul 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/mcp-goes-stateless-what-it-means-for-sre/https://cloudandsre.com/blog/mcp-goes-stateless-what-it-means-for-sre/The 2026-07-28 MCP release candidate deletes the session handshake for a stateless HTTP core and hardens OAuth. What changes for your agents, and when.Fri, 12 Jun 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/mcp-governance-linux-foundation/https://cloudandsre.com/blog/mcp-governance-linux-foundation/MCP now lives under the Agentic AI Foundation at the Linux Foundation. What vendor-neutral governance actually buys a platform team.Thu, 27 Aug 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/mcp-roadmap-2026-what-to-build-now/https://cloudandsre.com/blog/mcp-roadmap-2026-what-to-build-now/The MCP roadmap published 22 August 2026 names five priorities. Three change how you operate agent infrastructure — and you can prepare for all three today.Tue, 25 Aug 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/mental-models-for-ai-in-infrastructure/https://cloudandsre.com/blog/mental-models-for-ai-in-infrastructure/Tutorials answer how; mental models answer whether. Seven I use as the front gate before any LLM goes near a production system.Wed, 06 May 2026 00:00:00 GMTAISREArchitecturehttps://cloudandsre.com/blog/merkle-trees-and-content-addressing/https://cloudandsre.com/blog/merkle-trees-and-content-addressing/Derive the hash tree from the problem it solves, then find it everywhere — git, image digests, Certificate Transparency, and why a tag is not an identity.Mon, 05 Oct 2026 00:00:00 GMTFundamentalsSecurityhttps://cloudandsre.com/blog/model-eol-is-a-certificate-problem/https://cloudandsre.com/blog/model-eol-is-a-certificate-problem/Models retire on a schedule, the same model retires on different dates per platform, and 60 days' notice is a floor. Treat model EOL as inventory, not as news.Sun, 16 Aug 2026 00:00:00 GMTPlatform EngineeringAIReliabilityhttps://cloudandsre.com/blog/model-hardware-standard-field-brief/https://cloudandsre.com/blog/model-hardware-standard-field-brief/Anthropic's MHS gives agents read/write control of lab and factory hardware. What it guards, what it doesn't, and the controls you have to build yourself.Sun, 30 Aug 2026 00:00:00 GMTAIPlatform EngineeringReliabilityhttps://cloudandsre.com/blog/nas-vs-san-a-brief-comparison/https://cloudandsre.com/blog/nas-vs-san-a-brief-comparison/NAS serves files over ethernet; SAN serves blocks over a dedicated network. A brief comparison of the two storage architectures and when to pick each.Sun, 30 May 2021 00:00:00 GMTGeneralLinuxSREStoragehttps://cloudandsre.com/blog/nas-vs-san-for-gpu-workloads/https://cloudandsre.com/blog/nas-vs-san-for-gpu-workloads/File vs block was the old NAS-vs-SAN question. GPU training rewrote it. How the calculus shifts when storage has to keep an H100 cluster fed.Tue, 12 May 2026 00:00:00 GMTAIStorageInfrastructurehttps://cloudandsre.com/blog/no-anonymous-inference-endpoints/https://cloudandsre.com/blog/no-anonymous-inference-endpoints/The NSA and NIST put MCP on notice: agents are a funnel for prompt injection and privilege abuse. Why 'no anonymous inference endpoints' — and how to comply.Fri, 05 Jun 2026 00:00:00 GMTAISecurityPlatform Engineeringhttps://cloudandsre.com/blog/oauth-oidc-jwt-explained/https://cloudandsre.com/blog/oauth-oidc-jwt-explained/Three terms people use interchangeably and shouldn't: OAuth delegates access, OIDC proves identity, JWT is the token format. Mixing them up causes real bugs.Mon, 21 Sep 2026 00:00:00 GMTFundamentalsSecurityhttps://cloudandsre.com/blog/observability-and-incident-response-basics/https://cloudandsre.com/blog/observability-and-incident-response-basics/A primer on observability (logs, metrics, traces) and incident response (roles, severities, blameless postmortems) — the disciplines every SRE team runs.Sun, 24 May 2026 00:00:00 GMTSREObservabilityIncident ResponseFundamentalshttps://cloudandsre.com/blog/observability-for-ai-systems/https://cloudandsre.com/blog/observability-for-ai-systems/Golden signals miss the failure that pages you: a confident, well-formed, wrong answer. What AI observability adds — context as a span, quality as a signal.Fri, 05 Jun 2026 00:00:00 GMTAIObservabilityReliabilityhttps://cloudandsre.com/blog/okf-context-layer-for-ai-agents/https://cloudandsre.com/blog/okf-context-layer-for-ai-agents/The Open Knowledge Format gives agents a structured vocabulary for what data they're touching and where it came from — auditable, not blind, reasoning.Fri, 26 Jun 2026 00:00:00 GMTAIData EngineeringPlatform Engineeringhttps://cloudandsre.com/blog/open-weight-model-vm-to-100-customers/https://cloudandsre.com/blog/open-weight-model-vm-to-100-customers/An open-weight model serving HTTP takes twenty minutes. Making it something 100 customers can share takes a control plane. The whole path, with numbers.Thu, 13 Aug 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/owasp-agentic-top-10-for-platform-teams/https://cloudandsre.com/blog/owasp-agentic-top-10-for-platform-teams/OWASP's Agentic Top 10 reads like a security doc, but the mitigations are platform controls. Mapping all ten risks to infrastructure you already run.Thu, 09 Jul 2026 00:00:00 GMTAISecurityPlatform Engineeringhttps://cloudandsre.com/blog/owasp-mcp-top-10-audit/https://cloudandsre.com/blog/owasp-mcp-top-10-audit/OWASP now has a Top 10 dedicated to Model Context Protocol. The list is useful — but the census behind it is the part that should worry your platform team.Sat, 29 Aug 2026 00:00:00 GMTAISecurityMCPhttps://cloudandsre.com/blog/owasp-q3-2026-agentic-exploits/https://cloudandsre.com/blog/owasp-q3-2026-agentic-exploits/OWASP's Q3 2026 roundup reads as a containment failure report: agents escaped lab boundaries and reached real third parties. The prompt was the boundary.Thu, 08 Oct 2026 00:00:00 GMTAISecurityPlatform Engineeringhttps://cloudandsre.com/blog/percentiles-and-tail-latency/https://cloudandsre.com/blog/percentiles-and-tail-latency/Averaging percentiles across shards or minutes produces a number that means nothing. Here is the arithmetic, plus why p99 gets worse as you add services.Fri, 25 Sep 2026 00:00:00 GMTFundamentalsObservabilityhttps://cloudandsre.com/blog/postmortems-for-agent-caused-incidents/https://cloudandsre.com/blog/postmortems-for-agent-caused-incidents/Your incident template assumes a deterministic system and a human decision-maker. Agent incidents have neither. Five sections that fix the template.Sun, 02 Aug 2026 00:00:00 GMTSREAIReliabilityhttps://cloudandsre.com/blog/preparing-for-aws-solution-architect-professional-certification/https://cloudandsre.com/blog/preparing-for-aws-solution-architect-professional-certification/How I prepared for the AWS Solutions Architect Professional exam — study plan, resources, and what I'd do differently after passing.Tue, 30 Mar 2021 00:00:00 GMTAWSCertificationhttps://cloudandsre.com/blog/progressive-delivery-canary-and-rollback/https://cloudandsre.com/blog/progressive-delivery-canary-and-rollback/Every deployment strategy answers one question: who pays for the bug? Rolling, blue-green, and canary answer differently — only one limits blast radius.Mon, 17 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/prompt-engineering-for-sre/https://cloudandsre.com/blog/prompt-engineering-for-sre/Prompt advice is written for chatbots; SRE workloads are different. Six patterns I've shipped to production for SRE LLM tools, and why each earned its place.Tue, 05 May 2026 00:00:00 GMTAISREPrompt Engineeringhttps://cloudandsre.com/blog/prompt-injection-explained/https://cloudandsre.com/blog/prompt-injection-explained/An LLM sees one stream of tokens, not instructions and data. That is why prompt injection has no parser fix — and why provenance, not filtering, is the control.Tue, 22 Sep 2026 00:00:00 GMTFundamentalsSecurityAIhttps://cloudandsre.com/blog/queueing-theory-for-sres/https://cloudandsre.com/blog/queueing-theory-for-sres/Latency doesn't rise with load, it rises with 1/(1−utilisation). Two formulas explain the p99 cliff and size every pool in your system.Mon, 24 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/queues-and-message-brokers/https://cloudandsre.com/blog/queues-and-message-brokers/A queue decouples producers from consumers and absorbs bursts. Backpressure, at-least-once delivery, idempotency, DLQs — now in front of every LLM call.Sun, 18 Jan 2026 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/raft-consensus-explained/https://cloudandsre.com/blog/raft-consensus-explained/Consensus sounds academic until etcd loses quorum and your cluster freezes. Raft explained: leader election, log replication, and why majorities are the trick.Mon, 21 Sep 2026 00:00:00 GMTFundamentalsDistributed Systemshttps://cloudandsre.com/blog/rag-for-runbooks-on-call-knowledge/https://cloudandsre.com/blog/rag-for-runbooks-on-call-knowledge/On-call agents are only as good as what they retrieve. Making runbook RAG fresh, exact-match aware, and poison-resistant — with a runnable starter repo.Tue, 14 Jul 2026 00:00:00 GMTAISREOpen Sourcehttps://cloudandsre.com/blog/rate-limiting-algorithms-explained/https://cloudandsre.com/blog/rate-limiting-algorithms-explained/Five algorithms, one trade-off: burst tolerance vs memory vs precision. Plus why LLM APIs limit tokens instead of requests.Sun, 06 Sep 2026 00:00:00 GMTFundamentalsPlatform EngineeringAIhttps://cloudandsre.com/blog/recursion-and-the-call-stack/https://cloudandsre.com/blog/recursion-and-the-call-stack/Every recursive call costs a stack frame, and runaway depth ends in overflow. Frames, base cases, and why agent loops share the failure mode.Sun, 09 Mar 2025 00:00:00 GMTFundamentalsDevhttps://cloudandsre.com/blog/reliability-framework-autonomous-sre-agents/https://cloudandsre.com/blog/reliability-framework-autonomous-sre-agents/An autonomous airline agent rebooked 1,247 passengers wrong in one weather event. Trusting agents is a reliability problem — here's how to measure it.Mon, 06 Jul 2026 00:00:00 GMTAIReliabilityArchitecturehttps://cloudandsre.com/blog/rest-vs-graphql-apis-how-to-choose/https://cloudandsre.com/blog/rest-vs-graphql-apis-how-to-choose/REST or GraphQL? How the two API styles differ in data fetching, versioning, caching, and tooling — and how to choose for your service.Thu, 08 Dec 2022 00:00:00 GMTAPIDevhttps://cloudandsre.com/blog/retries-backoff-jitter-circuit-breakers/https://cloudandsre.com/blog/retries-backoff-jitter-circuit-breakers/A retry is a load multiplier. Backoff spreads it, jitter decorrelates it, a breaker stops it — and agent loops break all three.Fri, 04 Sep 2026 00:00:00 GMTFundamentalsReliabilityAIhttps://cloudandsre.com/blog/runner-is-the-api/https://cloudandsre.com/blog/runner-is-the-api/Everyone reviews the handler that restarts the pod. The safety lives in the runner around it — and handlers that skip it silently void every guarantee.Fri, 28 Aug 2026 00:00:00 GMTAIReliabilitySREhttps://cloudandsre.com/blog/sagas-vs-two-phase-commit/https://cloudandsre.com/blog/sagas-vs-two-phase-commit/Three services, one transaction, and no shared database. Why 2PC blocks, how sagas trade atomicity for compensation, and why compensation is not undo.Sun, 27 Sep 2026 00:00:00 GMTFundamentalsDistributed Systemshttps://cloudandsre.com/blog/schema-evolution-and-compatibility/https://cloudandsre.com/blog/schema-evolution-and-compatibility/Backward, forward and full compatibility — what each permits, why required fields are a trap, and the rules that keep a rolling deploy safe.Thu, 01 Oct 2026 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/secrets-rotation-and-blast-radius/https://cloudandsre.com/blog/secrets-rotation-and-blast-radius/Storing secrets safely is the easy half. The damage is decided by what one credential can do, for how long, and whether you can revoke it safely.Sun, 13 Sep 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/sharding-and-partitioning-explained/https://cloudandsre.com/blog/sharding-and-partitioning-explained/One database eventually runs out of one machine. Sharding buys headroom — but the shard key you pick is a near-permanent decision. Here's how to choose it.Mon, 21 Sep 2026 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/skills-for-agents-that-do-sre-work/https://cloudandsre.com/blog/skills-for-agents-that-do-sre-work/Most agent skills are chatbot prompts in disguise. Three operator-grade SRE skills — opinionated, output-contracted, portable across agent runtimes.Thu, 30 Apr 2026 00:00:00 GMTAISREOpen Sourcehttps://cloudandsre.com/blog/sli-slo-sla-error-budgets/https://cloudandsre.com/blog/sli-slo-sla-error-budgets/SLIs, SLOs, SLAs, and error budgets — the four numbers every SRE team must agree on, the math behind 'nines,' and what changes when agents burn the budget.Fri, 22 May 2026 00:00:00 GMTSREReliabilityFundamentalshttps://cloudandsre.com/blog/telemetry-integrity-trust-calibrated-autonomy/https://cloudandsre.com/blog/telemetry-integrity-trust-calibrated-autonomy/Autonomous remediation gates on model confidence and never on whether the telemetry is trustworthy. Telemetry integrity belongs in the gate too.Wed, 26 Aug 2026 00:00:00 GMTReliabilityAIObservabilityhttps://cloudandsre.com/blog/the-47th-approval-click/https://cloudandsre.com/blog/the-47th-approval-click/Human-in-the-loop is a real control until volume turns it into a reflex. How approval fatigue converts oversight into a rubber stamp — and the fix.Fri, 28 Aug 2026 00:00:00 GMTAIReliabilitySREhttps://cloudandsre.com/blog/the-7-layers-of-iso-osi-model/https://cloudandsre.com/blog/the-7-layers-of-iso-osi-model/The OSI model's seven layers give diverse systems a standard way to communicate. What each layer does and why the model still matters.Tue, 01 Jun 2021 00:00:00 GMTFundamentalsGeneralNetworkshttps://cloudandsre.com/blog/the-ai-native-sre-stack-2026/https://cloudandsre.com/blog/the-ai-native-sre-stack-2026/A practitioner's map of the AI-native SRE stack in 2026: six layers from telemetry to bounded remediation, and an honest read on where AI pays off.Fri, 12 Jun 2026 00:00:00 GMTAISREReliabilityhttps://cloudandsre.com/blog/the-cap-theorem/https://cloudandsre.com/blog/the-cap-theorem/Consistency, availability, partition tolerance — you get two. A walk through the CAP trade-offs, database by database.Wed, 02 Jun 2021 00:00:00 GMTFundamentalsGeneralStoragehttps://cloudandsre.com/blog/the-deployment-pipeline-build-once/https://cloudandsre.com/blog/the-deployment-pipeline-build-once/If your artifact is rebuilt per environment, staging never tested what production runs. Build once, promote everywhere is the rule the rest of CD rests on.Sat, 15 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/the-mcp-gateway-pattern/https://cloudandsre.com/blog/the-mcp-gateway-pattern/Letting agents call MCP servers directly repeats the no-API-gateway mistake. The five jobs an MCP gateway must do, with reproducible patterns for each.Sat, 02 May 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/the-memory-hierarchy/https://cloudandsre.com/blog/the-memory-hierarchy/Each memory level is 10–100× slower than the one above. Cache lines, locality, and why 'keep data near compute' is the biggest lever in LLM inference.Tue, 25 Jun 2024 00:00:00 GMTFundamentalsArchitecturehttps://cloudandsre.com/blog/the-reconciliation-loop/https://cloudandsre.com/blog/the-reconciliation-loop/Level-triggered control loops are the one idea behind Kubernetes, Terraform, GitOps, and every operator. See it once and half of platform work collapses.Mon, 10 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/the-slm-tier-model-right-sizing/https://cloudandsre.com/blog/the-slm-tier-model-right-sizing/Small models got good enough in 2026. But the win isn't using them — it's making model choice a platform tier with a router, eval gate, and demotion path.Sun, 02 Aug 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/the-verification-tax/https://cloudandsre.com/blog/the-verification-tax/DORA's 2026 ROI research names the verification tax and the J-curve. For platform teams the lesson is blunt: codegen moved the bottleneck, it didn't remove it.Fri, 14 Aug 2026 00:00:00 GMTPlatform EngineeringReliabilityAIhttps://cloudandsre.com/blog/timeouts-and-deadline-propagation/https://cloudandsre.com/blog/timeouts-and-deadline-propagation/A timeout is a local guess; a deadline is a shared fact. The difference decides whether your system sheds doomed work or grinds on abandoned requests.Wed, 19 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/tls-and-public-key-cryptography/https://cloudandsre.com/blog/tls-and-public-key-cryptography/Every https and model API call rides on TLS. How public-key crypto solves key distribution, what a certificate proves, and the cert-expiry realities.Sun, 10 Sep 2023 00:00:00 GMTFundamentalsSecurityhttps://cloudandsre.com/blog/token-exchange-delegation-vs-impersonation/https://cloudandsre.com/blog/token-exchange-delegation-vs-impersonation/When a service calls another service for a user, the token it carries decides whether your audit log says who acted. RFC 8693 makes that a design choice.Tue, 22 Sep 2026 00:00:00 GMTFundamentalsSecurityhttps://cloudandsre.com/blog/token-finops-the-third-budget/https://cloudandsre.com/blog/token-finops-the-third-budget/Error budgets, context budgets — agents add a third: dollars. Agent tasks burn 5–30× chatbot tokens, and cost-per-token is the wrong metric.Wed, 08 Jul 2026 00:00:00 GMTAIFinOpsSREhttps://cloudandsre.com/blog/tracing-the-agent-loop-otel-genai/https://cloudandsre.com/blog/tracing-the-agent-loop-otel-genai/Your agent is a distributed system wearing a chat interface. OpenTelemetry's GenAI conventions make it debuggable — what v1.41 covers and what's moving.Wed, 08 Jul 2026 00:00:00 GMTAIObservabilitySREhttps://cloudandsre.com/blog/types-of-agentic-memory/https://cloudandsre.com/blog/types-of-agentic-memory/Agentic memory is five things — working, episodic, semantic, procedural, entity — each with its own storage, eviction, and failure mode. A decision guide.Tue, 30 Jun 2026 00:00:00 GMTAIPlatform EngineeringArchitecturehttps://cloudandsre.com/blog/vector-index-hnsw-vs-ivf/https://cloudandsre.com/blog/vector-index-hnsw-vs-ivf/Exact search is O(n) and always right. Every ANN index buys speed with recall, memory, or build time. The arithmetic, compared.Tue, 08 Sep 2026 00:00:00 GMTFundamentalsAIDatahttps://cloudandsre.com/blog/virtual-memory-and-paging/https://cloudandsre.com/blog/virtual-memory-and-paging/Virtual memory is the beautiful lie every process lives in. Page faults, thrashing, overcommit — and why paging now manages the KV cache and OOM kills.Tue, 13 Aug 2024 00:00:00 GMTFundamentalsLinuxhttps://cloudandsre.com/blog/what-a-container-actually-is/https://cloudandsre.com/blog/what-a-container-actually-is/A container is not a lightweight VM. It is one ordinary process with three kernel restrictions applied. Knowing which three explains almost every container bug.Fri, 07 Aug 2026 00:00:00 GMTFundamentalsPlatform Engineeringhttps://cloudandsre.com/blog/what-are-vector-embeddings/https://cloudandsre.com/blog/what-are-vector-embeddings/A primer on vector embeddings — how meaning becomes something you can search, cluster, and compare, and the failure modes you only see in evaluation.Fri, 15 May 2026 00:00:00 GMTAIArchitecturePlatform Engineeringhttps://cloudandsre.com/blog/what-is-address-resolution-protocol-arp/https://cloudandsre.com/blog/what-is-address-resolution-protocol-arp/ARP discovers the MAC address behind an IP address. How the protocol works, the request/reply flow, and where it sits in the networking stack.Fri, 18 Nov 2022 00:00:00 GMTGeneralNetworkshttps://cloudandsre.com/blog/what-is-an-ai-agent/https://cloudandsre.com/blog/what-is-an-ai-agent/A primer on AI agents — the perceive-reason-act loop, what separates an agent from a one-shot LLM call, and the classical agent types SREs now operate.Mon, 11 May 2026 00:00:00 GMTAIAgentsPlatform Engineeringhttps://cloudandsre.com/blog/what-is-function-calling/https://cloudandsre.com/blog/what-is-function-calling/A primer on function calling — the JSON-schema contract that lets an LLM invoke your code. The request/response loop, parallel calls, and forced tools.Fri, 15 May 2026 00:00:00 GMTAIArchitecturePlatform Engineeringhttps://cloudandsre.com/blog/what-is-model-context-protocol/https://cloudandsre.com/blog/what-is-model-context-protocol/A primer on Model Context Protocol — the open standard that lets AI applications talk to tools through one interface. Hosts, clients, servers, transports.Mon, 11 May 2026 00:00:00 GMTAIMCPPlatform Engineeringhttps://cloudandsre.com/blog/what-is-prompt-caching/https://cloudandsre.com/blog/what-is-prompt-caching/Prompt caching cuts repeated-prompt cost 50–90% and halves latency. How prefix matching works, TTL economics by provider, and what decides your hit rate.Fri, 15 May 2026 00:00:00 GMTAIArchitecturePlatform Engineeringhttps://cloudandsre.com/blog/what-is-raid/https://cloudandsre.com/blog/what-is-raid/RAID provides redundancy for stored data, protecting against disk failures. The common RAID levels, how they differ, and when to use each.Sun, 30 May 2021 00:00:00 GMTLinuxStoragehttps://cloudandsre.com/blog/what-is-retrieval-augmented-generation/https://cloudandsre.com/blog/what-is-retrieval-augmented-generation/A primer on Retrieval-Augmented Generation — grounding an LLM's answer in documents you trust. Indexing, serving, and the failure modes that bite.Mon, 11 May 2026 00:00:00 GMTAIArchitecturePlatform Engineeringhttps://cloudandsre.com/blog/what-is-site-reliability-engineering/https://cloudandsre.com/blog/what-is-site-reliability-engineering/A primer on Site Reliability Engineering — where it came from, how it differs from DevOps and Platform Engineering, and what changes as AI joins on-call.Thu, 21 May 2026 00:00:00 GMTSREPlatform EngineeringFundamentalshttps://cloudandsre.com/blog/what-is-toil-and-the-50-percent-rule/https://cloudandsre.com/blog/what-is-toil-and-the-50-percent-rule/A primer on toil — the manual, repetitive work that eats SRE teams. Google's six-part definition, the 50% cap, and how AI agents change the playbook.Sat, 23 May 2026 00:00:00 GMTSREReliabilityFundamentalshttps://cloudandsre.com/blog/what-itbench-says-about-ai-sre/https://cloudandsre.com/blog/what-itbench-says-about-ai-sre/ITBench-AA put frontier models against 59 real Kubernetes incident diagnoses — all scored below 50%. What the benchmark measures and how to use it.Wed, 08 Jul 2026 00:00:00 GMTAISREReliabilityhttps://cloudandsre.com/blog/when-not-to-use-ai-in-production-sre/https://cloudandsre.com/blog/when-not-to-use-ai-in-production-sre/Most AI-for-SRE writing tells you where AI helps. Here are seven places it actively hurts — and the operational rule of thumb I use to decide.Sat, 25 Apr 2026 00:00:00 GMTAISREReliabilityhttps://cloudandsre.com/blog/why-ai-is-the-next-sre-superpower/https://cloudandsre.com/blog/why-ai-is-the-next-sre-superpower/After 15 years in cloud infrastructure and SRE, why I believe AI is the most significant shift in how we operate systems since Kubernetes.Mon, 20 Apr 2026 00:00:00 GMTAISRECloudhttps://cloudandsre.com/blog/why-your-ai-agent-should-not-act-yet/https://cloudandsre.com/blog/why-your-ai-agent-should-not-act-yet/Reading is reversible. Acting is not. Five things that must exist before an agent gets a write path — and the honest reason advisory-forever is also a failure.Fri, 28 Aug 2026 00:00:00 GMTAIReliabilitySREhttps://cloudandsre.com/blog/write-ahead-logging-explained/https://cloudandsre.com/blog/write-ahead-logging-explained/Pull the power on a database mid-transaction and it comes back intact. The trick is boring and universal: write your intent to a log before you touch the data.Mon, 21 Sep 2026 00:00:00 GMTFundamentalsStorage