Forward-Deployed AI Engineer — Remote/London, Equity

Deliverance AI; United States; Hybrid; USD 119,000 - 172,000; Full time

Job summary

Deliverance AI seeks a Forward-Deployed Applied AI Engineer to design and implement customer-specific agents and multi-agent workflows. You will ground workflows with customer data, connect knowledge sources, and govern telemetry and auditability as you iterate with users toward production-ready solutions.

You will balance output quality, latency, and cost while collaborating directly with customers in a remote-friendly, London-based role with flexible remote options.

Qualifications

• Hands-on experience building LLM or agentic applications that reached production.
• Strong software engineering fundamentals and the ability to deliver reliable systems.
• Practical discipline around prompting, testing and evaluation of AI behaviour.
• Comfort collaborating directly with customers and adapting to ambiguous requirements.
• Ability to balance quality, governance, latency and cost in real deployments.
• A finisher’s mindset focused on customer outcomes rather than prototype demonstrations.

Responsibilities

• Design and implement customer-specific agents, reusable skills and multi-agent orchestration.
• Ground workflows in customer information using retrieval-augmented generation and layered knowledge architecture.
• Tune routing and cost across workflows while preserving the required output quality.
• Instrument systems for operational telemetry, traceability and audit requirements.
• Work directly with customers to test, refine and productionise workflows under real constraints.
• Continue iterating until the delivered solution meets agreed quality and governance standards.

Skills

LLM applications

Software engineering

Prompting and evaluation

Job description

Deliverance AI seeks a Forward-Deployed Applied AI Engineer to design and implement customer-specific agents and multi-agent workflows. You will ground workflows with customer data, connect knowledge sources, and govern telemetry and auditability as you iterate with users toward production-ready solutions.

You will balance output quality, latency, and cost while collaborating directly with customers in a remote-friendly, London-based role with flexible remote options.