Enterprise Head of AI Engineering (Founding)Sales Development Representative (SDR)

About Pyxos

Pyxos is the agentic operating platform for data privacy compliance. Where incumbents sell workflow tracking tools and dashboards, we ship agents that execute compliance work — RoPA, DSARs, DPIAs, vendor risk, breach response, cross-border transfers — under human oversight. Our benchmark is an 80%+ reduction in manual privacy work. Pyxos was incubated at NEOM in Saudi Arabia. KSA is our first market; we expand from there across the GCC, then into the EU, UK, and US. We have one paying enterprise customer and six design-partner pilots launching with enterprise customers in KSA and the UAE. The founding team is senior. Pyxos is led by a six-time Silicon Valley venture-backed technology CEO with four significant exits, now based in London, alongside operating leaders with deep enterprise software and AI experience.

The position

You will own the technical direction of our agent surface, the proprietary build environment behind it (our Agentic Studio), and the evaluation, observability, and safety layers that make the system trustworthy enough for regulated enterprise deployment. We build with AI: agentic development tooling is core to how Pyxos ships, and we expect the person in this position to extend that approach, keeping the team deliberately lean. You will be part of the founding team, the most senior technical person, reporting to the CEO as part of the founding leadership team, and architectural authority over the AI layer. You will work closely with our VP, Product tightening the evaluation discipline to a standard that holds up under regulated scrutiny; and resolving the architectural decisions — model routing, memory, guardrails, audit trail — that define how the platform scales beyond v1. You will be writing code, not just directing it. Hiring is a parallel activity, paced to the work and held to a high bar; we would rather stay small longer than scale headcount ahead of need.

What we expect you to have done before:

• Shipped agentic systems in production. Multi-step LLM systems with tool use, sub-agent orchestration, persistent memory, and human-in-the-loop steps. You can describe specific failure modes and how you instrumented your way out of them.
• Evaluation as a discipline. You have designed and operated eval harnesses measuring faithfulness, hallucination rate, instruction-following, latency, and cost across model versions and prompt changes. For us, this is the clearest indicator of someone who has shipped LLM systems in production.
• Deep, current LLM engineering practice. System-level prompt design, RAG, structured generation, fine-tuning trade-offs (SFT, DPO, LoRA), and working knowledge of the frontier model landscape.
• LLM safety and security. Prompt injection mitigation, dual-LLM or plan-then-execute patterns, output validation, tool-use restrictions, policy enforcement.
• Production engineering rigor. Strong Python; cloud fluency (AWS, GCP, or Azure); CI/CD, observability, cost attribution.
• Engineering leadership at startup pace. You have hired, managed, and grown teams — not just been an individual contributor.

Nice to have:

• Regulated-industry experience (privacy, finance, healthcare, legal tech);
• BYOC or on-prem deployment experience;
• Familiarity with the EU AI Act or GDPR Article 22;
• Arabic language capability.

What we are not looking for

This is not a research role; we are not training foundation models or publishing papers as a primary output. We are building production systems where the cost of a wrong answer is a regulatory fine, which calls for a different set of disciplines and a different kind of engineer.

Logistics

Remote, UK or EU. Working pattern overlaps reasonably with European and Gulf time zones. Occasional travel to customer sites and team gatherings. Competitive base for an early-stage startup with meaningful founding-team equity; specifics in early conversations.

Process

Initial CEO conversation → technical deep-dive on agentic systems and evaluation → architecture working session on a Pyxos use case → references → offer. Three weeks, first conversation to decision. Pyxos is an equal opportunity employer. We hire on the basis of capability, judgment, and fit with what we are building. We welcome applications from candidates of all backgrounds and will make reasonable adjustments through the interview process on request.