Aidan Timmerman

Deployments and Partnerships @merge_api

[Alexandr Wang @alexandr_wang]

muse wonder full on merge’s intelligence index

[Aidan Timmerman @aidan_timmerman]

@shensi @merge_api The team at merge is unparalleled

[Aidan Timmerman @aidan_timmerman]

@danielyegarcia Congrats Daniel !!

[Google Cloud Tech @GoogleCloudTech]

Terminal test runs won't catch multi-turn agent regressions.

In Ep 3 of Agent Clinic, Dani Zamora & @matt_feroz build an automated eval suite for a LangGraph agent in 60 mins.

Here's the 4-step framework to go from vibes to benchmark ANY AI agent → https://t.co/MbQq33xBkl

[Aidan Timmerman @aidan_timmerman]

@prit4k @noahrshinn Ruh roh

[Grok @grok]

Merge’s Universal Context Layer is looking like a real contender for GOAT status in AI connectivity. It turns every employee’s AI lessons into shared company knowledge, cutting token spend while lifting accuracy for the whole org—like a franchise QB who makes the entire roster better every snap.

[Shensi Ding @shensi]

Introducing Merge’s Universal Context Layer.

It substantially decreases token cost while increasing accuracy.

Every company uses AI wrong:

Say your designer makes a deck using Claude. They teach Claude your design systems and typography.

Now a Marketing hire makes a deck. None of the information your designer taught Claude gets passed over… why?

Merge fixes this:

All knowledge and memory across your entire company gets stored inside our context layer.

When someone on the org teaches Al, that skill lives across the entire organization instead of staying locked in their session.

Your AI runs faster, cheaper, and with more accuracy. You don’t have to switch systems, incur more costs, or retrain anyone on your team, Merge runs invisibly in the background.

Book a demo here: https://t.co/aj3ecxk7zP

[Shensi Ding @shensi]

Token growth right now on Merge Gateway is insane.

Last week I posted that @merge_api token volume was up 11x month over month, and had to correct it to 14x a few hours later.

As of today, it's 37.8x.

[Aidan Timmerman @aidan_timmerman]

@nishkachotai @merge_api @pradbiswas3 Best team in the game!! If you want to join, Merge is hiring

[Shensi Ding @shensi]

really wild - token processed by @merge_api gateway has increased 11x month over month

[NEA @NEA]

Introducing Merge for Workforce!

📋 IT picks the model routing policies for each team, fast and cheap for routine work, frontier models for the hard stuff, etc.
📞 Every call ties back to a person, team, and model, so IT finally sees where the AI budget actually goes.
⚙️ IT is able to audit and grant access to approved connectors and skills straight to employees' AI tools, so people get the right access without asking, and nothing shows up that wasn't approved.
💻 The controls follow employees across whatever they use, Claude, ChatGPT, Cursor, etc.

Congratulations, @shensi @GilFeig & the @merge_api team! 👏🏽👏🏽👏🏽

More details below 👇🏽

[Aidan Timmerman @aidan_timmerman]

Merge has landed in San Diego

[Aidan Timmerman @aidan_timmerman]

@shensi What does Merge do? Merge created the AI solution every company has been asking for

[Shensi Ding @shensi]

We just cut AI spend by 75% across an entire workforce in one click.

Introducing: Merge for Workforce.

Without Merge, your employees are left choosing one of two options:

1. Spend 100x the cost they should be from using frontier models for everything

2. Deliver subpar work from overly restrictive model access

Instead with Merge, IT can connect any identity provider and set model routing policies by team in one click.

Merge pushes them to every machine through a desktop client, overriding the model configs across your team’s AI assistants and coding tools.

Now every task gets the right model for the job.

The result? 75x fewer tokens with faster and better output.

[Shensi Ding @shensi]

Energy is a constraint for model hosting, and it's going to make model routing with fallbacks an essential part of the ecosystem.

The grid can't add power fast enough, and the wait to plug a big new site in runs for years. So hosts build wherever they can get power, often on their own generation onsite. That capacity is cheaper and quicker to stand up, and it skips the redundancy that gets standard hosting to around 99.99% uptime. It runs closer to 99.9%. So the capacity growing fastest is also the least guaranteed.

Routing and model hosting grow together from here.

As more of this cheaper, less reliable capacity comes online, the way to run on it reliably is to spread your requests across several hosts and switch automatically when one goes down. That becomes the default way people build on AI models, and model routing with fallbacks turns into standard infrastructure.

[Aidan Timmerman @aidan_timmerman]

@Suhail @merge_api has one, and they're giving startups $5,000 in free spend https://t.co/voLKUadC6e

[Aidan Timmerman @aidan_timmerman]

@shensi @lennysan @merge_api Merge has a cracked eng team !

[Merge @merge_api]

Venture and YC startups have been flooding our inboxes, especially about Gateway.

Today, we're introducing the Merge Startup Program! Thousands of FREE credits for early-stage teams to start building.

If you're an early stage startup we want you building with Merge Gateway.

Try it out: https://t.co/hr8WvPWg7G

[Max Eagle @MaxDEagle]

Over the last month, I've spent a lot of time working on and thinking about AI evals. Here are some of the harder, less obvious lessons:

[Aidan Timmerman @aidan_timmerman]

@shensi @jemmamerge @merge_api Providing safe transit for the loaf today