Skill is a leading technology company that connects top talent with major brands. They are seeking a visionary Software Engineer to design and implement a centralized AI platform that enhances developer productivity and streamlines operations by integrating AI capabilities into development workflows.
Responsibilities:
- Architect and implement sophisticated agent orchestration systems, including router functions, queueing mechanisms, and agent invocation services
- Seamlessly integrate AI agent capabilities into modern CI/CD platforms, designing pipeline configuration patterns where agent invocations execute as stages, process context from code changes, produce artifacts, and control subsequent stages based on AI agent outcomes
- Develop and expand a suite of shared, tool-based functions that empower AI agents with versatile capabilities
- Design and implement advanced knowledge bases, leveraging powerful search and storage technologies to ensure AI agents can reason over a comprehensive, shared organizational context rather than isolated prompts
- Implement robust AI guardrails, including input/output filters, sensitive data scrubbing, content moderation, and fine-grained, per-agent permission boundaries, ensuring secure and compliant AI operations by design
- Develop and integrate TokenOps controls for efficient resource management, encompassing model tiering and intelligent routing via an LLM gateway, semantic caching, and optimized context-window management
- Instrument the entire AI platform with comprehensive observability, including dashboards, audit trails, and detailed trace logging for every agent invocation (input, output, decision, tokens, cost) to ensure compliance, facilitate debugging, and provide deep operational insights
Requirements:
- 3–7 years of professional experience in software development
- Strong proficiency in Python and the Python SDK for building production-grade serverless functions and event-driven services
- Solid experience with cloud-based services, including serverless compute, API management, messaging queues, event buses, identity and access management, secrets management, and monitoring tools
- Specific experience with AI agent services, action groups, knowledge bases, and guardrails
- Hands-on experience integrating with large language model (LLM) APIs, such as those from leading providers or similar platforms
- Familiarity with CI/CD platforms and best practices
- A strong security mindset, including expertise in least-privilege access, secure secrets handling, input validation, and an awareness of prompt injection and data-leak risks in LLM workflows
- Comfortable with observability principles, including structured logging, metrics, tracing, and writing highly debuggable code for production environments
- Experience with vector search and Retrieval-Augmented Generation (RAG), including vector databases, embeddings, and retrieval evaluation
- Familiarity with FinOps or TokenOps concepts, such as cost attribution, intelligent model routing, semantic caching, and batch inference
- Experience building developer platforms or internal tooling that other engineers rely on daily
- Familiarity with advanced AI agent frameworks and an understanding of their architectural tradeoffs