Lenovo is seeking to advance its Hybrid AI vision and make Smarter Technology for All by designing, developing, and exploring next-generation AI systems.
Requirements
- Strong programming experience in Python.
- Solid CS fundamentals: data structures, algorithms, complexity, concurrency.
- Experience building and debugging non-trivial software systems (class projects, internships, research code, or open-source).
- Coursework or project experience in machine learning or deep learning (e.g., supervised learning, optimization, neural networks).
- Familiarity with at least one modern ML framework such as PyTorch, TensorFlow, or JAX.
- Conceptual understanding of large language models (LLMs) or other foundation models (what they are, common use cases, limitations).
- Experience using LLMs via APIs (e.g., OpenAI, Anthropic, etc.) or open-source models for a project, internship, or research.
Responsibilities
- Build agentic AI workflows Design and implement multi-step AI agents that can plan, call tools/APIs, reason over intermediate results, and complete complex tasks end-to-end.
- Develop orchestration layers Implement services that route and coordinate calls between foundation models, tools, data sources, and other agents (e.g., planners, executors, evaluators).
- Design context and memory management Help design and implement strategies for managing context and “memory” across interactions (e.g., chunking, retrieval, long-context handling, and stateful workflows), so agents can work over large and evolving information.
- Prototype and productionize Take ideas from notebook-level prototypes to robust, observable, production-ready systems, including APIs, services, and internal tools.
- Work with foundation models Fine-tune, adapt, and post-train large models (e.g., instruction-tuning, RAG, tool-use finetuning, preference optimization) in collaboration with senior ML engineers.
- Experiment and evaluate Design experiments, build evaluation harnesses, and analyze metrics to compare different prompts, policies, and agent strategies.
- Engineer for reliability and safety Contribute to guardrails, monitoring, and fallback strategies so agentic systems behave predictably and safely in real user environments.
Other
- Bachelor’s degree in Computer Science, Computer Engineering, or a closely related technical field (e.g., EE, Math, Statistics, Physics) with a strong computing/programming focus.
- Ability to work effectively in a team setting: code reviews, design discussions, and cross-functional collaboration.
- Clear written and verbal communication skills.
- This position is NOT ELIGIBLE FOR VISA SPONSORSHIP, including Optional Practical Training (OPT) or Curricular Practical Training (CPT).
- All applicants must be currently authorized to work in the United States for any employer.
