Applied AI Engineer at Seamflow

London, United Kingdom

on-site · full_time · Visa sponsorship

Apply for this role at Seamflow

Responsibilities:
- Design, build, and ship LLM-powered features end-to-end: define problems, craft prompts, implement integrations, deploy, and iterate.
- Build reliable evaluation pipelines and monitoring to detect regressions and guide model improvements.
- Design and implement multi-step agents, tool usage, orchestration, and function-calling for complex workflows.
- Implement document intelligence: extract structured data from PDFs, scans, forms, and tables; handle messy real-world inputs.
- Work directly with customers: review real documents, debug failures, and prioritize product fixes.
- Collaborate across frontend (TypeScript/React) and backend (Python) to embed AI into real user workflows.

Requirements:
- Strong Python skills and hands-on experience with modern LLM tooling (OpenAI/Anthropic/Azure SDKs), structured outputs, and function calling.
- Experience building production AI systems (beyond simple chat wrappers) with monitoring, CI/CD, and maintainability in mind.
- Practical experience in prompt engineering, building meaningful evals, and diagnosing model failure modes.
- Ability to reason about when to use which model and how to make model behavior reliable in production.
- Comfortable working across the stack (frontend and backend) and shipping product-facing features from Day 1.
- Startup/early-stage experience, founder background, or exposure to ambiguous requirements is a strong plus.