Founder & CEO @zerodrift_ai · The Enforcement Runtime for AI · a16z @speedrun Scout · Founded Accern (acquired)
New York City
- Tweets 206
- Following 141
- Followers 950
- Likes 136
Introducing Anchor 3.0 by @zerodrift_ai, the first enforcement model built to secure AI agents and keep everything they say compliant at runtime.
It caught more compliance violations than every frontier model we tested, up to 34x faster and up to 12x cheaper.
Live today, first 1,000 calls free: zerodrift.com
658
16
7,392
279,825
Built for regulated industries, open to anyone. Sign up, grab an API key, and your first 1,000 calls are free. After that it's $0.01 per message with no minimums and no sales call.
zerodrift.com
1
1
6
235
None of this ships without the @zerodrift_ai team. Proud to build alongside every one of you.
Thank you to @HelloSurgeAI for the benchmark partnership, and to @speedrun, @andrewchen, @Tocelot, @tkexpress11, @kenanhsaleh, @patmatthews, @koenchristiaens, @Pitchdrive_vc, @mahaniok, @ConvergeVC and everyone who backed us early.
1
10
238
The calls for an AI slowdown are getting louder.
Our biggest launch of 2026 drops tomorrow at 10am ET. It's the reason AI doesn't have to.
Live for everyone.
29
2
15
10,739
Your AI said it. You own it.
Seven months ago @zerodrift_ai launch. This week we're in Times Square.
We chose those six words because they're the truth the AI industry is avoiding. When your AI speaks for your company, you carry the liability.
So we built the fix. Every AI message gets checked and fixed before it sends. Compliance, enforced.
Thank you to our partners at @brexHQ for putting us up there.
zerodrift.com
3
2
12
722
Kumesh Aroomoogan retweeted
Across 150+ seed rounds this year, $855M in funding has gone toward companies building at the intersection of AI and security.
As agents move out of the sandbox and into production, runtime enforcement is becoming essential to deploying them successfully, especially in regulated industries.
Excited to see ZeroDrift’s $10M seed round included in @Crunchbase’s latest recent roundup.
news.crunchbase.com/cybersec…
1
2
3
456
Today we're launching Guard for Agents: the enforcement runtime for AI 🔥
Your agent: "Full refund approved, plus $500 for the trouble!"
No such policy exists. You're on the hook. Now multiply by 10,000 conversations a day.
Guard catches it, names the policy, and ships this instead:
"Your order is past our 30-day refund window, but I can offer store credit or a replacement."
Milliseconds, before delivery.
32
21
161
25,364
Guard ships just three weeks after Command. Everyone in security is racing to control what AI can access and do, but the thing a customer experiences is the message. That layer is now enforced.
Docs: docs.zerodrift.com Workspace + $1,000 in credits: tony@zerodrift.ai
Thanks to our supporters at @speedrun, @andrewchen, @Tocelot, @tkexpress11, @kenanhsaleh, @patmatthews, @koenchristiaens, @Pitchdrive_vc, @mahaniok, @ConvergeVC, and many more.
1
5
219
Live now:
1/ MCP connectors for Notion, Confluence, Drive, OneDrive + Linear: policy docs → enforceable rules
2/ Validation API: one call in your response path + OpenAPI spec for typed clients
3/ Anchor 3.0-Preview, our compliance SLM + LoRA adapters for your policies (preview)
4
367
Kumesh Aroomoogan retweeted
We're opening hiring, again, for 2 roles in NYC.
- Senior Backend Engineer
- Senior Product Marketing Manager
We're a growing AI Lab that builds our own models. SLMs, LLMs, Training Studios, and one powerful platform to run it on.
Come join us!
DMs are open.
job-boards.greenhouse.io/zer…
ZeroDrift raised an oversubscribed $10,000,000 seed round to enforce compliance on AI communications, just 3 months after launching out of stealth.
AI agents are becoming the next enterprise communication layer, and every regulated company will need compliance enforced before AI writes, speaks, calls, or publishes on its behalf.
TechCrunch exclusive: techcrunch.com/2026/06/02/ze…
7
3
56
5,255
Companies are putting AI in front of customers faster than they can control it.
That’s why we built @zerodrift_ai.
Every AI message is checked against your policies and regulations before it goes out. Pass, rewrite, or block. Every decision logged.
Last week we launched Command, our control plane for the enforcement runtime, and generated 1M+ impressions across the developer community.
You set the rules. Your AI can’t break them.
10
13
25
1,532
Kumesh Aroomoogan retweeted
ZeroDrift has moved from .ai to .com.
A small change, easy to miss.
But when it comes to enforcing policy across millions of AI-generated communications, the smallest mistakes can have a big impact.
Good thing catching them is kind of our thing.
Find us at zerodrift.com to learn more.
2
1
3
243
The agent-native workplace will require a new set of compliance tools for both humans and agents - @kumesh_a has been building the command center for this
Today we're launching @zerodrift_ai Command ⚡️
Set company policies, turn on regulation rulepacks, and enforce compliance on every AI communication before it's sent. Blocked and corrected in milliseconds.
Plus: Anchor 3.0-Preview, the first compliance enforcement SLM (limited onboarding).
Sign up at zerodrift.com to get a Command workspace + $1,000 in complimentary credits.
10
3
26
4,479
Today we're launching @zerodrift_ai Command ⚡️
Set company policies, turn on regulation rulepacks, and enforce compliance on every AI communication before it's sent. Blocked and corrected in milliseconds.
Plus: Anchor 3.0-Preview, the first compliance enforcement SLM (limited onboarding).
Sign up at zerodrift.com to get a Command workspace + $1,000 in complimentary credits.
46
26
105
61,605
ZeroDrift Command comes just 2 months after our $10M seed. The round was a bet that every enterprise will need compliance enforced before AI speaks on its behalf. Today that's live.
Thanks to our supporters from @speedrun, @andrewchen, @Tocelot, @tkexpress11, @kenanhsaleh, @patmatthews, @koenchristiaens, @Pitchdrive_vc, @mahaniok, @ConvergeVC, and more
1
2
8
640
Today, @zerodrift_ai cleared the full slate on security and compliance.
ZeroDrift is the compliance layer for AI communications. We check messages against your regulatory obligations and internal policies before they’re sent.
We are now:
🔒 SOC 2 Type II
🔒 ISO 27001:2022
🔒 GDPR compliant
🔒 HIPAA compliant
We started this work before our first customer call. Because when your platform sits inline with regulated AI communications, security isn't paperwork. It's the product.
Thanks @TrustVanta for the partnership!
zerodrift.ai/company/securit…
14
16
48
14,486
1/ Excited to announce that I’m an a16z Scout for the a16z @speedrun Fund, investing in the next wave of AI founders from day one 🚀
If you’re building something ambitious, reply below with a one-line description of what you’re building and I'll get back.
264
27
450
208,767
2/ Over the last 13 years, I’ve built enterprise AI companies, sold Accern, raised $70M+, and most recently founded ZeroDrift.
a16z Speedrun backed ZeroDrift before incorporation, when it was just an idea. That early conviction changed our trajectory from day one.
Since then, Speedrun has helped us hire exceptional talent, build enterprise pipeline, and raise an oversubscribed $10M seed in just 3 weeks.
Now I’m looking to back AI-native founders from pre-seed through seed, especially across infrastructure, developer platforms, and enterprise applications.
I can write checks within 24 hours, back you before incorporation, and help create visibility with the Speedrun team.
Feel free to send what you're building to kumesh@aroomoogan.com.
6
2
20
2,006