2 months ago
London, United KingdomSenior / Staff+
Responsibilities
- Design and build agentic workflows and AI-powered features for Elliptic’s copilot using frameworks such as LangChain or LangGraph.
- Lead the design of agents, tool integrations, retrieval pipelines, and multi-step AI workflows for blockchain investigation data.
- Build and own evaluation frameworks for LLM output quality, reliability, latency, and regression testing.
- Design and build backend services, APIs, and event-driven systems using TypeScript and Node.js.
- Lead technical design reviews and architecture decisions for AI workstreams.
- Mentor and coach junior engineers through pairing, code review, and design feedback.
- Improve engineering practices for testing, observability, AI system reliability, prompt management, and rollout safety.
- Influence how AI capabilities integrate with the broader Elliptic platform and ship significant AI capabilities end-to-end.
Requirements
- 5-8+ years of software engineering experience, including at least 1-2 years of meaningful production experience building LLM-powered features.
- Experience with LLM frameworks such as LangChain, LangGraph, or similar.
- Hands-on experience building agentic systems involving tool use, multi-step reasoning, planning, memory, and human-in-the-loop patterns.
- Practical understanding of prompt engineering, structured outputs, context management, and LLM trade-offs.
- Experience building and running LLM evaluations, including evaluation sets, LLM-as-judge approaches, and regression testing.
- Strong backend skills in TypeScript and Node.js with solid API design experience.
- Cloud experience with AWS services such as Lambda, ECS, and S3, or similar platforms.
- Proficiency with SQL and Postgres plus some NoSQL experience.
- Demonstrated ability to mentor and coach other engineers.
- Experience using AI coding assistants such as Copilot, Cursor, or Claude critically and effectively.
- Preferred experience with retrieval-augmented generation, vector databases, multi-agent systems, agent orchestration, LLM observability tools, multiple model providers, fine-tuning, Terraform, Kubernetes, infrastructure-as-code, Datadog, SNS, or SQS.
- Interest in cryptocurrency, blockchain, or compliance.
Benefits
- Hybrid working with the option to work from almost anywhere for up to 90 days per year.
- £500 remote working budget for home office setup.
- $1,000 learning and development budget.
- 25 days of annual leave plus bank holidays and an additional birthday day.
- Sixteen weeks of fully paid enhanced parental leave for eligible employees.
- Private health insurance through Bupa, Spill Mental Health Support, life assurance, and Cycle to Work Scheme.
Tech Stack
Categories
