Skip to main content
PilotGet a free AI cost & reliability audit: every fix priced in dollars per month. You keep the savings; I get a case study, a testimonial and your feedback.2 of 3 places leftHow the pilot works
Ankit Jha

Why there are no client case studies yet

I'd rather show nothing than show something I can't back up. The pilot program is open: 3 free audits, in exchange for an anonymized case study, a short testimonial and a 30-minute feedback call. The first write-up, LLM Bill Teardown #1, follows the first pilot audit.

Apply for a pilot place
case_study.template

Every case study will follow the same shape

  1. 01A summary box that stands alone for skimmers
  2. 02Context
  3. 03Problem
  4. 04Constraints
  5. 05Diagnosis
  6. 06Decisions and trade-offs
  7. 07Outcome in $ and eval pass rate
  8. 08Lesson

Systems I've built

Production systems from before the consulting practice, anonymized where needed: the problem, the architecture, the evals and the outcome.

The first write-ups are being prepared. Until they're up, email me and I'll walk you through them on a call.

What will appear here

Flagship cases first, then supporting work. Each card leads with one $ number.

  1. Next

    LLM Bill Teardown #1

    Flagship

    Where a SaaS team's model spend actually went, and which fixes held quality. Published with the client's approval.

  2. Next

    Open-source eval gate

    Open source

    A small GitHub Action that runs a golden eval set on every pull request and blocks changes that drop the pass rate.

  3. After launch

    What this site costs to run on AWS

    Experimental

    A small, real cost breakdown of a static site.

AI Engineering Economics

Articles on building LLM systems that hold up in production, and what they cost. Each one by email as it goes live; nothing else.

Articles are published on Hashnode.

Get new articles by email