Blog & articles

Long-form engineering writing across LinkedIn, guest posts on Avinteli (cloud security & governance), and TrueFoundry (AI platform engineering). Each card opens the canonical article on the publisher site.

• LinkedIn

  3 hidden AWS cost drains nobody talks about

  How Inter-AZ and NAT data processing, CloudWatch Logs ingest and endless retention, and orphaned EBS volumes or snapshot chains quietly eat 30–40% of spend in otherwise “healthy” stacks—and what to measure first. Includes a personal cost-and-reliability dashboard built on the free tier (Lambda, S3, CloudWatch, Terraform).

  Read

• LinkedIn

  Why I’m Rethinking DevOps: From Delivery Speed to Business Impact

  A shift from pipeline velocity to provable value: cost as a design input, reliability as a contract (SLOs and error budgets), and security as operational integrity—grounded in real projects from AI/ML cost work to EKS SLOs, AWS assessments, and multi-region GitOps.

  Read

• LinkedIn

  From Panic to Procedure: Designing Runbooks and SLOs that New Engineers Can Actually Use

  Operational knowledge for 2 a.m. incidents: symptom-first runbooks, concrete checks and escalation paths, user-centred SLOs paired with error budgets, and how pairing runbooks with SLOs builds trust and faster onboarding.

  Read

• LinkedIn

  Automate or Fall Behind – Streamlining Vulnerability Management in Modern AWS Cloud Environments

  Why periodic scans and manual patching fail at cloud speed; continuous detection and remediation with DevSecOps; AWS Security Hub, Inspector, Config, Lambda, and tools like Trivy in CI/CD; compliance as ongoing enforcement, not a yearly audit.

  Read

• LinkedIn

  Seamless AWS Authentication with GitHub Actions: A Practical Guide for Real-World CI/CD Using Terraform

  Replace long-lived AWS keys in GitHub with OIDC: register GitHub as an IAM OIDC provider in Terraform, scope trust to repo and branch, attach least-privilege policies, and assume a role from workflows with aws-actions/configure-aws-credentials.

  Read

• Avinteli

  Simplify Cloud Governance with AWS Control Tower

  Control Tower as governance-by-design: landing zones, account factory, guardrails and Config against drift, and why multi-account sprawl needs encoded policy—not wiki pages. Includes a realistic enterprise onboarding narrative ahead of SOC 2–style assurance.

  Read

• TrueFoundry

  Claude Code Workflow: How It Works and How to Use It in Production

  How agentic coding loops actually run: context ingestion, tool calls, feedback, and iteration; where workflows break (context loss, weak traces, flaky tools); and production habits—prompt constraints, scoped permissions, checkpoints, and instrumentation.

  Read

• TrueFoundry

  Claude Code MCP Integrations: How Tools Connect to AI Coding Agents

  Model Context Protocol for coding agents: tool discovery and call_tool flows, types of integrations (files, Git, APIs, logs), limits of “native” MCP for governance and observability, and why gateways and tight tool scoping matter in production.

  Read

• TrueFoundry

  Kong vs LiteLLM: Architecture, Pricing, and Trade‑Offs

  Two philosophies for AI traffic: enterprise API management extended to LLMs versus a thin, provider-normalising LLM proxy—setup cost, governance depth, and where operational gravity lands. Frames managed gateways as a middle path for teams outgrowing ad hoc proxies.

  Read

← Home