A normal Lambda function that has to wait pays for the wait. If your workflow extracts a document, then waits five minutes for a confirmation, the function sits there billing for all five minutes, and anything longer than the 15-minute limit does no...
!Image descriptionhttps://dev-to-uploads.s3.us-east-2.amazonaws.com/uploads/articles/c5smdlluse3wn13p9tf1.png
For a decade, 900 seconds was the hard wall on a single Lambda invocation. If one continuous computation needed longer, Lambda was the wron...
🛡️ One Line in a Pod Spec, a VM Underneath: A Practical Guide to AWS Nitro Sandboxes on EKS
> TL;DR
> - Pods on a node normally share one Linux kernel. A container escape or kernel bug in one pod can reach its neighbors. Nitro Sandboxes give each o...
I recently moved a structured-extraction pipeline from Claude Sonnet 4.6 to Sonnet 5.5 on Amazon Bedrock. On paper it is a model ID change: the same family, a lower price per token, better benchmark scores. In practice, some requests that worked on 4...
Invoices from the field rarely arrive as clean PDFs. They arrive as phone photos: the invoice lying on a stack of other papers, a coffee receipt on top of it, or a sheet held against a car dashboard. I wanted to know which extraction setup holds up i...
🧠 Spatial Memory vs Correction Memory: I Benchmarked Both Against a Plain LLM
> TL;DR
> - Two open-source memory layers for LLM document extraction, measured on the same 120 synthetic invoices and forms, next to a no-memory baseline, with Claude Ha...
🤖 From Writing the Spec to Approving It: What AI-DLC Looks Like on a Real Build
> TL;DR
> - AI-DLC is public and open source: awslabs/aidlc-workflowshttps://github.com/awslabs/aidlc-workflows. The method comes from this AWS DevOps blog posthttps://...
🌐 Organization-Wide EC2 DNS Automation with EventBridge and AWS Config Aggregator: One Rule Per Account
> TL;DR
> - Detecting EC2 launches and terminations across a whole AWS Organization for DNS is not the five-minute single-account job it looks l...
🛡️ Kubernetes Security Is Three Problems, Not One: Admission, Identity, Runtime
> TL;DR
> - Most Kubernetes compromises start inside the perimeter, where the firewall verifies nothing.
> - Treat security as three independent problems: admission wha...
🛡️ A Field Guide to Beating Amazon Bedrock Throttling
> TL;DR
> - Know your quota first: check Service Quotas, learn which limits are adjustable soft and which are not hard.
> - Retry with exponential backoff and full jitter on throttling errors. T...
🔑 The Bottleneck Was Not the Model. It Was the Prompt Lookup.
> TL;DR
> - A managed prompt-management service was throttling prompt retrieval on the hot path, which capped the whole pipeline.
> - The model was never the bottleneck. Fetching the pro...
🌉 The GitOps Bridge: Keep Terraform and ArgoCD Out of Each Other's Way
> TL;DR
> - Infrastructure Terraform and applications ArgoCD/Flux have different change frequencies, approval flows, and blast radii. Coupling them is the root problem.
> - The ...
🚦 Your Readiness Probe Is Lying to You
> TL;DR
> - KEDA's Metrics Server reported Ready even when its gRPC connection to the keda-operator was down.
> - In that state the HPA silently stops receiving scale signals, while the replica still looks hea...
🛡️ A Prompt-Injection Firewall for AWS: What I'd Build, and What I Measured
> TL;DR
> - Prompt injection is OWASP's #1 LLM risk across every edition, and a 2026 proof shows pure input-wrapper defenses face a hard trilemma: they cannot be continuous...
After OCR, most pipelines throw the layout away and re-discover the same form from scratch every time. textract-field-memoryhttps://github.com/aws-samples/sample-textract-field-memory is a small, zero-dependency Python layer that remembers where fiel...
> TL;DR — I open-sourced an AWS sample that makes LLM document extraction learn from human corrections. Each QA fix is captured once and reused, so the pipeline gets more accurate and cheaper over time — with no retraining and no redeployment. Repo: ...
A field survey of the Kubernetes + LLM inference stack in 2026 — the problems that now have serious players, and the edges that are still genuinely underserved.
If you run large language models on Kubernetes, you've probably noticed the ground shift...
Most production LLM extraction pipelines route every field through a frontier model, every time. It works, and for a while nobody questions the bill. Then volume grows, or someone runs the per-document math, and the real question shows up. How much o...