Job Description

AI Engineer

Focus: Anthropic Claude & AI-Powered Application Development

Role Overview

As an AI Engineer, you are responsible for the design, development, and operationalization of AI-powered solutions built on Anthropic's Claude. Working within cross-functional agile teams, you will architect and implement applications that leverage large language models (LLMs) to solve real-world business problems. Your work spans the full development lifecycle — from translating business requirements into prompt strategies and system architectures, through to deployment, monitoring, and continuous improvement of AI systems in production. You will take end-to-end ownership of AI features and be a key contributor to the organization's AI engineering best practices.

Responsibilities & Tasks

Requirements & Solution Design

Collaborate with product owners and agile team members to understand business requirements and translate them into AI-driven solutions
Evaluate use cases for Claude suitability and design effective prompt architectures, RAG pipelines, and agentic workflows

Prompt Engineering & Model Integration

Design, implement, and iterate on prompts and system instructions to achieve reliable, high-quality model outputs
Develop and maintain integrations with the Claude API, including multi-turn conversations, tool use, and structured output generation
Apply advanced prompting techniques such as chain-of-thought reasoning, few-shot learning, and context management strategies

Application Development

Build production-grade AI applications and services (REST APIs, backend services, AI agents) that integrate Claude into existing or new systems
Implement retrieval-augmented generation (RAG) architectures, including document ingestion pipelines, vector database management, and semantic search
Develop tool-use and function-calling integrations that allow Claude to interact with external systems and APIs
Ensure robustness through error handling, fallback strategies, output validation, and graceful degradation

Quality, Testing & Operations

Design and implement evaluation frameworks to benchmark model performance, accuracy, and consistency across use cases
Write unit, integration, and end-to-end tests for AI-powered features, including LLM output validation strategies
Monitor deployed AI systems for quality, latency, cost, and safety; implement alerting and observability tooling
Contribute to CI/CD pipelines for AI applications, including automated evaluation and regression detection

Safety, Ethics & Governance

Apply responsible AI principles in system design, including safety filtering, bias evaluation, and alignment with Anthropic's usage policies
Identify and mitigate AI-specific risks such as prompt injection, hallucination, and data leakage in production systems
Stay current with Anthropic's model updates, safety guidelines, and best practices, and propagate this knowledge within the team

General Competence Description

Professional with proven expertise in:

Prompt engineering and LLM application development, including system prompt design, context management, tool use, and multi-turn conversation handling
Software engineering fundamentals: RESTful API development, clean code practices, version control (Git), and CI/CD pipelines in at least one major language (Python preferred)
AI/ML ecosystem familiarity: vector databases, embedding models, RAG architectures, and LLM evaluation frameworks
Responsible AI: understanding of AI safety concepts, output validation, risk mitigation strategies, and ethical deployment considerations

Additional Competence Description

5+ years of software engineering experience, with at least 1 year working directly with LLM APIs (Claude, OpenAI, or equivalent) in production settings. Demonstrated ability to take AI features from prototype to production, including performance optimization, monitoring, and iterative improvement. Experience in an agile environment with strong communication skills to bridge technical and non-technical stakeholders.

Nice to Have

Hands-on experience with Anthropic's Claude API, including use of extended thinking, and the Messages API
Familiarity with LLM orchestration frameworks such as LangChain, LlamaIndex, or similar
Experience with cloud platforms (AWS, Azure, GCP) and containerized deployments (Docker, Kubernetes)
Background in the insurance, finance, or another regulated industry
Contributions to open-source AI tooling or published evaluations/benchmarks

Confidential — Internal Use Only