At a Glance
- Tasks: Architect and build agent scaffolding for enterprise-grade security and develop evaluation infrastructure.
- Company: Helmguard.ai is a pre-Series A startup in London, backed by top investors from Meta and SpaceXAI.
- Benefits: Significant pre-Series A equity upside and a culture of radical ownership.
- Other info: Collaborate with labs like Anthropic to resolve core engine bugs.
- Why this job: Join a high-growth startup achieving seven-figure revenue shortly after launch.
- Qualifications: Proven backend engineering experience in TypeScript with 1-2 years of production-grade LLM features.
The predicted salary is between 70000 - 90000 £ per year.
Helmguard.ai is a pre-Series A London startup building agent-native risk infrastructure, backed by top UK/US investors and alumni from Meta, SpaceXAI, and Palantir. Founded by Palantir and Stanford/Oxford experts, they have reached seven-figure revenue within months of launch and secure multi-year contracts with global financial and healthcare enterprises.
You will own the agent platform, transforming frontier model calls into production-grade enterprise features for high-stakes risk management. As a founding engineer, you’ll build the orchestration, evals, and reliability infrastructure that allows AI agents to act as peers to domain experts, setting the standard for AI quality and safety at scale.
Location: London, UK
Why this role is remarkable:
- Join a high-growth startup that achieved seven-figure revenue months after launch, backed by tier-one investors and tech heavyweights from SpaceXAI and Palantir.
- Experience the outsize impact and influence of a founding-team role, with significant pre-Series A equity upside and a culture of radical ownership.
- Work at the extreme frontier of AI, pushing APIs so hard you'll collaborate with labs like Anthropic to resolve core engine bugs.
What You Will Do:
- Architect and build agent scaffolding including tool use, context management, sandboxing, and robust prompt-injection defenses for enterprise-grade security.
- Develop sophisticated evaluation infrastructure for high-stakes outputs, utilizing LLM-as-judge frameworks and regression testing to ensure peer-level correctness.
- Engineer reliability systems including custom retries, circuit breakers, and prompt versioning to turn experimental model outputs into dependable production actions.
The ideal candidate:
- Proven backend engineering experience in TypeScript with at least 1-2 years of shipping production-grade LLM features and multi-step agent orchestration.
- Strong systems thinking regarding asynchronous queues, idempotency, and the ability to curate datasets for evaluating fuzzy, high-stakes compliance policies.
- A self-starter comfortable owning AI quality end-to-end, possessing the technical conviction to say 'no' when features don't meet rigorous safety bars.
Founding Engineer, Agent Systems at Helmguard employer: Jack & Jill
Helmguard.ai, located in London, is pioneering agent-native risk infrastructure with significant backing from top-tier investors. The team comprises experts from Palantir and Stanford/Oxford, ensuring a cutting-edge approach to AI safety and quality.
