About Hamming
Hamming is the security layer for voice and chat agents. Banks, health systems, and other enterprises are putting agents on phone lines where anyone can reach them, and where the agent can verify identity, read an account, and act. We attack those agents before criminals do, with the owner's permission, show what they fall for, and keep every fix tested on every release.
We are one of the fastest engineering teams in the world. We prod deploy 4x / day.
OpenAI Dev Day 100billion token list
The role
Our researchers find the failure by hand. You make it happen a thousand times, against authorized targets, with evidence a customer can inspect, and again on every release. You own the harness: the runner that places calls and chat sessions, the evaluators that decide what a result means, and the replay that shows what changed. Thousands of concurrent calls, bounded, cancellable, observable.
What you'll do
Build the attack harness. A new technique is a new definition, not a new system.
Generate useful variations. Branch multi-turn scenarios and voice conditions with the objective and conditions explicit.
Make results trustworthy. A successful attack, a refusal, an infrastructure failure, and an inconclusive result are distinguishable and inspectable.
Make findings reproducible. Scenario, configuration, model versions, tool events, outcomes, replay, and comparison.
Keep fixes fixed. Validated findings become regression tests that run on every change, with noise controlled.
What you'll bring
3+ years building backend services, async workers, or test infrastructure in Python or TypeScript, with the failure cases and the tests.
An adversarial mind: offensive tooling, CTFs, fuzzing, or a security project. The title does not matter.
Judgment about evidence: when an automated evaluator is enough and when a human has to look.
Idempotency, backpressure, observability, and limits that are enforced rather than documented.
Nice to have
Agent frameworks, voice or telephony systems, or LLM evaluation infrastructure
A security finding of your own, responsibly disclosed
How we work
We start with the customer problem, ship daily, and let evidence decide what comes next. Use AI tools to move faster and own the result. Direct, low ego, serious work.
Benefits and perks
Competitive salary and meaningful equity
Health, dental, and vision coverage
Every frontier AI tool, fully paid, from day one
Conferences and events we sponsor
Flexible PTO
Offices in San Francisco and Austin; remote in the United States for most roles