Free 30-minute intro call, no commitment. Fixed-price CDK review from $495, agreed before any paid work starts. See how pricing works Your team's time: a walkthrough of your goals and one of the report; we do the rest.
A linter finds style issues. We look at how the app is put together: which stack owns what, how environments are wired and how code reaches production. Then we turn the fixes into reusable, best-practice constructs, so the right way becomes the default way for every stack that follows.
We use AI to widen the search, then apply production CDK judgement to decide what is actually risky, what should become a construct, and what should stay boring by design.
My project became so big that claude can't properly understand it
QuestionSo, I made a project in python entirely using Cursor (composer) and Claude, but it has come to a point that the whole codebase is over 30 Python files, code is super disorganized, might even have duplicate loops, and Claude keeps forgetting basic stuff like imports at this point. When I ask it to optimize the code or to fix a bug, it doesn't even recognize the main issue and just ends up deleting random lines or breaking everything completely.
I have 0 knowledge about python, it's actually a miracle I got this far with the project, but now it's almost impossible to keep track of things, what do I do? already tried using cursor rules but doesn't seem to work.
A real post from r/ChatGPTCoding. Code nobody structured stops fitting in anyone's head, the model's included.
The engagement is built to move quickly. We are not delivering an abstract slide deck. We are reviewing the repo, identifying the structural issues that matter, and handing back changes your team can use.
We start from your goals, pain points and current repo structure. Read-only access to the repository is enough; we never need AWS credentials.
We review how the app is put together: stack boundaries, environments, constructs, deployment flow, reuse and security, looking for what gets harder to change as the team grows.
You get an architecture report with prioritized recommendations, and a pull request that introduces best-practice constructs for the most important fixes.
We review your CDK app's architecture the way a senior reviewer would and walk your team through what slows every change down. You leave with a report and a pull request that introduces best-practice constructs for the first fixes: build on them yourselves, or have us refactor the rest for a fixed quote. Larger codebases take longer to review, so $495 is where it starts.
Finding. Buckets are defined inline in 4 stacks with the same settings, and reports-stack.ts has already lost enforceSSL.
Recommendation. Move the settings into a construct under lib/constructs and keep only what differs in each stack.
“Danny helped review my CDK app and provide feedback. He turned around the work very quickly and created a detailed document for me to follow. His document was thorough and included concrete, easy to follow examples.”
Pick it when the accounts and guardrails around the CDK app need rebuilding.
See the AWS CDK Landing ZonePick it when you want the running workload reviewed, not only the code.
See the AWS Well-Architected ReviewWe'll talk through your repo, where CDK changes slow the team down, and whether the review is the right first step. If it isn't, we'll say so.
You talk to Danny Steenman, the engineer who does the work. Pick a time below and he confirms it by email.