Lower AI bills, same quality
same quality
To make every AI request cost as little as it can without hurting the answer, by sitting with each team and building a proxy tuned to how it really works.
Our mission
People
Who does what for you
We do the work with you, not hand you a tool. Every engagement has the same roles behind it: someone to read your bill, someone to build and tune the proxy, someone to check quality never drops, and someone to keep the savings coming as usage changes. These are roles, not named people; you meet the engineers on the first call.
Takes the first call and goes through one month of your usage: which tools, teams and models cost what, where the waste is, and the saving to expect. Your point of contact from the first call onward.
Sits with your engineers, learns your tools, agents and workloads, then builds the proxy around them: what to trim, what to cache, which model takes which task. Stays on it until the saving is real.
Reviews your spend by model and task, then writes the rules that trim context, send easy work to cheaper models, cache repeated context and queue jobs that can wait for batch pricing.
Tests every rule on your own work before it goes live and compares the answers with what you get today. Keeps up with what each model does well and what it costs.
Owns the month-to-month relationship: reports spend and savings by team, schedules rule reviews as new models and prices arrive, and makes sure changes land without breaking anyone's workflow.
Works with your finance lead on the baseline, the monthly report of spend and savings, and budgets with alerts and hard stops so no month can surprise anyone.
Sends every person their key and a short guide for their tool, whether that is Claude Code, Codex, Cursor, an SDK or your own agent harness, and checks the first requests go through.
First response for day-to-day questions: a new joiner, a leaver to revoke, a cap to raise, a rule to adjust, a tool that will not connect. Keeps a record of every change made on your account.
why proxen
Why we built Proxen
AI bills keep climbing. Coding agents and API calls are billed per token, and most teams send every request to the strongest model, resend long histories on every turn, pay for the same context again and again, and run work that could wait at full price. None of that shows up on the invoice. It just shows up as a bigger number every month.
Gateway products give you the parts and leave the tuning to you. We think the saving is in the tuning. So we sit with your team, learn your whole setup, build a proxy for it and keep tuning it as models, prices and usage change. Same models, same tools, same quality, a much smaller bill. Proxen is built and run by Surehand (surehand.io).
How we work
Our principles
Cheaper, never worse
A saving only counts if the answers stay as good. We test every change on your own work, and the log shows which model answered every request.
Honest numbers
Up to 70-80% is what a tuned proxy can do on the right workload. Your number comes from your own usage, measured against a baseline we agree before we start.
Your accounts, your data
Everything runs on your company's own API accounts, inside the providers' terms. Prompts are not stored by default. Leave whenever you like and nothing breaks.
Find out what your AI bill should be
Send us one month of usage. We show you where the money leaks and what you can expect to save, for free. Then we pilot on one team.
