БразилияSenior Data Scientist - Generative AI
2 нед. назад
БразилияSenior
llmdiffusionprompt engineeringvector retrievalragproductionization
Lead development and production of generative AI systems and retrieval pipelines for a live creative-generation platform.
О компании
- Created in 2022, the Growth Group unifies our business-to-business (B2B), direct-to-consumer (DTC), Sales & Distribution, and Marketing teams. By bringing together global tech and commercial functions, the Growth Group allows us to fully leverage data and drive digital transformation and organic growth for AB InBev around the world.
- In addition to supporting well known global beer brands like Corona, Budweiser and Michelob Ultra, the Growth Group is home to a robust suite of digital products including our B2B digital commerce platform , on-demand delivery services Ze Delivery and TaDa Delivery, and table top beer keg PerfectDraft.
- We are an exceptional team, focused on understanding and supporting consumer and customer needs, harnessing new technology, and scaling growth opportunities.
Обязанности
- We're hiring a Senior Data Scientist to help lead our GenAI platform — the systems behind prompt-driven solutions and the retrieval/RAG pipelines that ground them in real product and brand data. You'll own both the generative side (LLM and diffusion-based creative pipelines) and the search side (embeddings, vector retrieval, RAG performance) of a production system serving live creative-generation traffic, and help set technical direction for how the team builds and evaluates GenAI and search systems going forward.
- Design, build, and productionize GenAI systems end to end: from prompt engineering to the RAG architectures (retrieval, ranking, and generation) that ground them in accurate, brand-safe outputs.
- Own the search and retrieval layer for GenAI applications: embedding models, vector similarity search, and re-ranking strategies, continually evaluating retrieval quality (precision/recall, relevance) and latency/performance, and iterating on chunking, indexing, and hybrid search techniques.
- Stay up to date with the latest advancements in LLMs, multimodal generation, and retrieval-augmented generation. Apply frontier techniques (fine-tuning, few-shot/in-context learning, agentic and multi-step RAG) to solve complex creative and search problems, driving innovation across the team's GenAI roadmap.
- Utilize cloud and orchestration platforms (Azure, Azure OpenAI/Foundry, Databricks, Spark) to process large-scale datasets and serve GenAI and RAG systems efficiently and at production scale through async APIs and background job pipelines.
- Utilize Python (FastAPI, PySpark) for data manipulation, embedding pipelines, and modeling.
- Implement automation and evaluation workflows: offline eval, A/B testing, monitoring of retrieval and generation quality to streamline iteration and enhance productivity.
- Collaborate closely with cross-functional teams, including data engineers, ML/platform engineers, business stakeholders, and product managers, to understand GenAI and search requirements and deliver high-quality, production-grade solutions. Embrace an agile development approach to iterate quickly and efficiently, and mentor other data scientists on GenAI and retrieval best practices.
Требования
- Bachelor's degree in computer science, engineering, mathematics, or another quantitative field. A master's degree or PhD is a plus.
- Relevant experience in data science or applied machine learning, with strong technical skills and a deep understanding of frontier GenAI techniques: LLMs, multimodal (text-to-image) generation, prompt engineering, and RAG architectures.
- Hands-on experience with search and retrieval systems: embedding models, vector/similarity search, hybrid (semantic + keyword) search, and re-ranking, along with a strong grasp of retrieval performance: latency, throughput, and recall/precision trade-offs, and how to evaluate and tune them in production.
- Experience with LLM/embedding APIs (OpenAI, Azure OpenAI, or similar). Experience implementing and deploying generative and retrieval models using these frameworks.
- Experience working with cloud platforms such as Azure (Azure OpenAI/Foundry, Blob Storage, Event Hub), Databricks, and Spark for big data processing, model serving, and scalable GenAI system design.
- Proficiency in Python/PySpark for data manipulation, analysis, and modeling tasks. Strong knowledge of API frameworks (e.g., FastAPI) for productionizing models as services.
- Good knowledge of CI/CD tools like GitHub for version control and collaboration, and of testing/typing tooling (pytest, mypy, linting) for shipping reliable, production-grade GenAI services. Understanding of workflows and automation tools, including async job orchestration and evaluation pipelines, to streamline processes and enhance efficiency.
Условия
- Performance based bonus*
- Attendance Bonus*
- Private pension plan
- Meal Allowance
- Casual office and dress code
- Days off*
- Health, dental, and life insurance
- Medicines discounts
- WellHub partnership
- Childcare subsidies
- Discounts on Ambev products*
- Clube Ben partnership
- Scholarship*
- School materials assurance
- Language and training platforms
- Transport allowance
- *Rules applied