Hey, I'm Carl.

I’m a Software Engineer with 11 years of experience. I help AI SaaS companies optimize their LLM bills through my own “Guard, Embed, Cascade” system.

Where other consultants look for ways to make your LLM bills cheaper through minor tweaks, I look for ways to remove portions of your LLM bill entirely, without sacrificing product quality.

Case Studies

Case Study: Cutting LLM Costs by 90% Without Losing Quality

I run Kuverly, a social listening tool that ingests huge volumes of posts and surfaces only the relevant few to customers. That relevance judgment ran through a multi-shot LLM pipeline — several passes per post, tuned hard for quality. This is the process I used to cut our own LLM bill

Carl AndersonCarl Anderson

How it works

“Guard, Embed, Cascade” is all about reducing the total number of LLM calls that happen in the first place, as well as optimizing LLM calls themselves to be as cheap as possible while still meeting quality requirements.

This causes profit margins to move from a high cost-per-user caused by LLMs towards the flat cost structures that enabled traditional SaaS companies to enjoy massive profit margins.

• Guard: Protect LLM calls with circuit breakers and traditional ML to completely avoid unnecessary LLM calls.
• Embed: Embeddings are a powerful tool, much cheaper than regular LLM costs and are reusable. Change LLM calls to use embeddings when performing tasks such as routing, search, and inference.
• Cascade: When LLM calls are unavoidable, cascade to the minimum viable model to keep costs minimal.

Find out what your LLM bill should actually cost

For LLM bills over $5000 a month, reach out to me to discuss your setup.

I can help you find:

• The true cost of your AI bills
• A ranked breakdown of opportunities to improve your high-cost LLM flows
• Places where unnecessary LLM calls are eating away at your margins