• Understand customer needs through AI: Build models that interpret intent and context to provide the right product, recommendation or experience at the right moment. • Fine-tune and adapt LLMs: Apply LoRA/qLoRA and RAG techniques to customize state-of-the-art LLMs for Gopuff’s domain, using frameworks such as Hugging Face, LangChain and LlamaIndex. • Architect personalization systems: Design robust pipelines for search, retrieval and ranking that combine generative AI with classical ML, leveraging vector databases and prompt/context optimization. • Drive algorithmic innovation: Explore and apply new methods in deep learning, reinforcement learning, multi-task learning and embeddings to improve personalization, recommendations and discovery. • Develop conversational and Q&A experiences: Build interactive agents that can answer questions, guide product discovery and engage with customers in natural language. • Deploy at scale: Collaborate with MLOps and engineering teams to productionize models with high scalability, reliability and low-latency response times. • Evaluate and optimize: Define metrics, design A/B tests and perform offline/online evaluations to measure model performance, customer satisfaction and business impact. • Collaborate and educate: Work with product, engineering, design, analytics and leadership teams to translate business needs into AI solutions and communicate data-science concepts clearly. • Mentor and lead: Provide technical leadership and mentorship to data scientists and ML engineers, fostering a culture of responsible, customer-focused AI. • Stay ahead of the curve: Continuously track emerging LLM architectures, tools and techniques, and integrate them into Gopuff’s AI roadmap.