Neon Postgres’ Post

We sat down with the team at Neon to discuss how we manage tens of thousands of databases for our users at Specific: https://lnkd.in/dBR8smJ6 Being able to spin up scalable Postgres databases for our users in seconds has been magical and a big enabler in realising our vision of an entire cloud optimized for agents. We have been growing our usage of Neon fast and feel very confident that Neon will keep scaling with us. Neon being "just Postgres" is key to us. Your coding agent can write its code around Postgres and use Specific to provision and connect it securely, both locally and in production. Shifting this process left lets the agent not just write code but also autonomously test it and build confidence that it will work once deployed. A simple "specific deploy" is then all it takes to get scalable, production infrastructure. Specific might be the simplest way to get a Neon database running together with all the other infrastructure you need for any app!

Cloud infrastructure like AWS and Cloudflare did away with the need to manage physical servers, patch operating systems, and maintain uptime. Instead of managing infrastructure, developers could focus on their application logic while the cloud provider handled the rest. Devin applies the same principle to coding agents: the sandbox, dev environment, agent runtime, orchestration, integrations, and tooling are all fully managed. You bring the prompt; Devin handles everything behind it. https://lnkd.in/gTS2uWa6

What is Terraform? -> Terraform is an infrastructure-as-code tool that lets you build, change, and version cloud and on-prem resources safely and effectively. Use of Terraform: -> Terraform is an infrastructure as code tool that lets you define infrastructure resources in human-readable configuration files that you can version, reuse, and share. -> You can then use a consistent workflow to safely and efficiently provision and manage your infrastructure throughout its lifecycle. 👍 👍 👍

Modern networks have steadily moved from command‑line interfaces to cloud‑managed dashboards, but in doing so, some of the strengths of traditional text‑based configuration have been lost (where has my CLI gone?) Many networks are now configured via a GUI - but there is real power in programming ability in the network. Read our blog 'The Network as Code: Network Programmability,' here: https://lnkd.in/g76a5tzJ

Built my first custom MCP server today. Actually had tried previously a couple times and somehow it went sideways. But finally got it. Next I'll push it to the cloud so whole team can access. That said the pre-built connectors in Claude are useful, but they didn't cover what I specifically needed for marketing and ops workflows. So I used Claude to help me write the code and build my own. If you haven't looked into MCP servers yet, worth it. You can extend Claude's capabilities well beyond the default toolset/connecters available.

Your cloud is only as secure as your weakest IAM policy. And I guarantee you have at least one that would make me smile during a red team. 'Just give it admin' during development becomes permanent in production. Overprivileged IAM roles show up in every single cloud assessment we perform. The attack paths: - Leaked access keys in git repos found and exploited within hours - SSRF to the instance metadata service stealing IAM role credentials - Cross-account pivoting through misconfigured trust policies - Lambda functions with admin access giving maximum blast radius Found AWS access keys in a public GitHub repo that gave full admin to a production account with 200+ running resources. Pivoted from development to production through one misconfigured line in a cross-account role trust policy. The cloud shared responsibility model means the platform is secure. Your configuration is your problem. When was the last time you audited IAM policies for overprivileged access? If it's been more than 90 days, it's overdue.

Add Real GCP Control Use GCP APIs to start/stop VMs, manage storage, deploy apps (not just redirect). Backend Security Layer Node.js APIs to handle OAuth tokens securely. Role-Based Access (RBAC) Admin/User roles with restricted permissions. Live Monitoring Show CPU, usage, billing via Cloud Monitoring API. Result: Converts project from UI clone → real full-stack cloud management system. part-1

Whether you want to learn how to architect a complex multi-agent system with long term memory, leverage local and remote MCP servers for tool standardization, or write detailed Terraform scripts for secure Cloud Run deployment, this four part blog series will show you how. In this first part of our series, we'll focus on establishing the agent's core capabilities by standardizing its external integrations through the MCP → https://goo.gle/4sjTLXV

If you’ve ever built an app and it worked on your laptop but broke on another machine… Docker solves that problem. So what is Docker? Docker is a tool that lets you package your app together with everything it needs (libraries, runtime, configs) into a container. A container is like a self-contained mini-environment that runs exactly the same on any machine — your laptop, a server, or the cloud. No more environment issues. No more “but it works on my system.” Some benefits of using docker - Consistency Your app runs the same everywhere. - Fast Setup Spin up a ready-to-run environment in seconds. - Lightweight Uses fewer resources than virtual machines. Docker helps you build once and run anywhere — cleanly, reliably, and efficiently.