AI Engineering Intern (LLM Ops & RAG)
hace 7 días
Palma
LearnWise is a company where the best idea always wins — no matter what — and where innovation and hyper-growth are in our DNA. We offer Engineering internships designed to build future technical leaders - hopefully at LearnWise! Internship · EU · Remote Department: Engineering Reports to: Head of AI Work alongside: CTO, AI team, Engineering & Product Job Description LearnWise.ai is a scale-up modernizing educational institutions with virtual assistants, instructor efficiency, and student engagement solutions for higher education. We’re seeking an inspired and rigorous Engineering Intern for a 6-month internship focused on AI systems, reliability, and product behavior. You’ll join a VC-funded team with a track record of exponential growth. You’ll work closely with leadership and be expected to propose, test, and ship your own ideas. What you’ll do AI Systems & Reliability Troubleshoot LLM behavior and agent actions across our Tutor Assistant, automatic feedback, and other AI features; identify root causes and ship fixes. Use LangSmith, LogFire and other tools to trace requests end-to-end, correlate prompts, tools, and outputs, and explain why a behavior occurred. Implement pragmatic guardrails, fallbacks, and caching; reduce failure modes and improve correctness under real usage. Retrieval & Relevance Optimize our retrieval pipelines: choose and tune embeddings, refine vector-database indexing/search strategies, and evaluate re-rankers for higher precision/recall. Design and run experiments (e.g., recall@k, MRR, nDCG, groundedness checks) that translate directly into product wins. LLM Ops & Tooling Contribute to LLM Ops: prompt/tool versioning, dataset curation, regression test suites, latency/cost monitoring, and safe rollout strategies (shadow, A/B). Build small utilities to surface failure taxonomies, detect drift, and turn traces into actionable fixes. Grow into feature development that measurably improves product quality and student outcomes. Collaboration & Strategy Partner with the Head of AI, CTO, AI team, Engineering, and Product to shape technical direction. Share learnings clearly; propose changes that improve model performance, reliability, and UX. Requirements Core Requirements Strong proficiency in Python. Master’s (ongoing or completed) in AI/ML or a related field, or equivalent experience demonstrating comparable depth. Solid understanding of modern AI models and AI-powered applications (prompting, tool-use/agents, RAG, context management, evaluation). Hands-on experience with LLMs and/or tool-using agents: debugging prompts, tracing tool calls, analyzing failure modes. Ability to design evaluation loops (golden sets, regression tests, online experiments) that make reliability measurable and improvable. Ability to figure things out independently and chase issues until completion. Nice to Have Experience optimizing RAG components: embeddings selection/tuning, vector DB configuration, re-ranker selection/evaluation. Familiarity with AI observability/monitoring practices and failure taxonomies. Fluent in English; direct, candid communication style; hungry to learn and ship. What’s in it for you? Direct mentorship from senior leadership (CTO, Head of AI, Head of Product). Work with a global team on cutting-edge AI applications. Flexible working arrangements (results > hours). A “no 9-to-5 mentality” and genuine focus on work-life balance. If you’re passionate about engineering with AI, eager to learn, and excited to work in a fast, practical startup environment, this is your chance to build AI that meaningfully improves student outcomes. We’d love to hear from you.
APLIDECOR BALEARActivo hace 4 horas
