AI Software Engineer (LLM & Automation)

Job Description

• Design and build AI-powered tools to improve software engineering and operational workflows across multiple product teams.
• Collaborate with technical leads and engineers to understand priorities, codebases, architecture, deployment, and engineering toil sources.
• Translate needs into practical AI solutions adopted by teams.
• Work hands-on in applied AI domains: developer productivity, workflow automation, retrieval and knowledge systems, agentic workflows, intelligent engineering tools.
• Ensure systems are evaluated, observable, reliable, and secure for daily engineering use.
• Communicate effectively to gather requirements, explain trade-offs, negotiate scope, and document decisions.

Required Skills & Qualifications

Must-have

• 5-7 years of experience overall, 3+ years relevant.
• Strong software engineering experience in production apps, developer tools, automation, internal platforms, or data-intensive services.
• Proficiency in Python, TypeScript, Java, or similar.
• Experience delivering LLM-powered applications.
• Experience with retrieval-augmented generation, agentic AI workflows, AI developer/operational tools, workflow automation, AI evaluation/monitoring.
• Integration experience with Git, CI/CD, issue trackers, documentation, logging, feature flags, messaging.
• Strong understanding of APIs, testing, debugging, observability, authentication, error handling, secure delivery.
• Ability to learn unfamiliar systems by reading code and collaborating.
• Experience managing ambiguous problems through design, implementation, validation, rollout, adoption.
• Strong communication skills for technical collaboration.
• Ability to work independently and manage multiple priorities.

Nice-to-have

• Experience with agent frameworks (LangGraph, OpenAI Agents SDK, LangChain).
• Knowledge systems (semantic/hybrid retrieval, embeddings, sources like Git, Confluence, Jira).
• Code intelligence/search (source-code indexing, dependency graphs, OpenSearch, Elasticsearch).
• Agentic system design (planning, tool use, memory, retries, human approval).
• Developer tooling and coding agents (Claude Code, Cursor, GitHub Copilot).
• Tool integration and automation (function calling, APIs, event-driven workflows).
• Evaluation and observability (synthetic evaluations, regression testing, LLM-as-judge, LangSmith, OpenTelemetry).
• Familiarity with AI models and security (Anthropic, OpenAI, Amazon Bedrock, guardrails, secure data handling).
• Systems and infrastructure (deployment using containers, APIs, workflow engines).
• Proven track record shipping AI tools impacting productivity, quality, reliability, cost, or operations.