×
Manifold
AI detection and response platform that watches what your AI agents actually do at the endpoint, in real time
Last verified August 15, 2026 · Updated daily
What Manifold is building
Manifold is a security platform for AI agents on enterprise endpoints. The problem it solves is specific: coding agents like GitHub Copilot, Claude Code, and Cursor already sit on developers' laptops with access to source code, production systems, CI/CD pipelines, and an expanding network of MCP servers. Those agents take real actions, executing shell commands, making API calls, reading entire codebases, and today's endpoint detection and response tools (EDR) are blind to them, or treat all of that activity as suspicious and issue constant false positives. Manifold gives security teams visibility into what AI agents are actually doing at runtime: which agents are deployed, what they access, what actions they take, and whether any of that behaviour is anomalous. Think of it as an EDR built specifically for the agent era, rather than retrofitting 2018 endpoint security for 2026 agent behaviour.
Why this matters
The timing of this raise is not accidental. AI agent adoption in enterprise has crossed a threshold in the last six months: 85% of developers reportedly use coding agents regularly. That figure is about to extend to every knowledge worker as general-purpose agents (Claude Cowork, OpenClaw, and others) reach their desks. The security gap this creates is structural and not going away: traditional EDR tools were designed to detect malware, not to understand whether a legitimate AI agent has been compromised or manipulated. The founders' specific insight, that developers already get blanket exceptions to endpoint security policies because their normal activity looks malicious, is exactly right, and it means the "assume the agent is trustworthy" posture is baked into enterprise security culture by default. Manifold's bet is that as agent activity spreads from developers to every knowledge worker, that posture becomes untenable. The angel investors tell the story clearly: Joe Sullivan built the security function at Uber through one of the most high-profile corporate data breach episodes in tech history, and Vijay Bolina ran CISO at Google DeepMind, two people who think professionally about what happens when AI systems do things organisations don't expect. Their cheques are a specific thesis, not just financial support.
Open roles at Manifold
3 positions we're tracking. Roles are re-checked daily and removed when filled.
Security Research Engineer
First seen 4 months ago
Software Engineer (Agent Monitoring Infrastructure)
First seen 4 months ago
Enterprise Security GTM / Solutions Engineer
First seen 4 months ago
Working at Manifold
Since , Manifold has built AI detection and response platform that watches what your AI agents actually do at the endpoint, in real time. The team is now <50 people. Working at a B2B, Supply Chain and Logistics company at this stage means broad remit, direct access to founders, and equity that still means something if the company works out.
The majority of roles are in Seattle, WA, USA; Remote.
Frequently asked questions
How many jobs does Manifold have open?
3 open roles at Manifold, last checked March 2025.
Does Manifold hire remotely?
Manifold doesn't have remote openings at the moment. All roles are in Seattle, WA, USA; Remote.
What roles is Manifold hiring for?
Manifold is hiring across Engineering. The most recent opening is Security Research Engineer.
How do I apply for a job at Manifold?
Use the apply links above, or check our guide to getting hired at Manifold.
Where is Manifold based?
Manifold is headquartered in Seattle, WA, USA; Remote.
Get Manifold roles before they're posted
We track B2B, Supply Chain and Logistics companies like Manifold daily and surface roles before they reach the job boards. Fewer applicants, faster replies, real hiring managers.
Get early access →