GenAI & ML Data Scientist — Hybrid (Finance AI)

Calibrate People; Sandton; Hybrid; ZAR 900,000 - 1,300,000; Full time

Job summary

Calibrate People in Sandton, Johannesburg, SA, is seeking a Data Scientist with deep GenAI/ML expertise to design, build, and deploy AI-powered applications at scale. You will work across backend and frontend layers, and implement MLOps to support reliable deployments.

The role requires strong knowledge of embeddings, vector search, NLP, and Python, with collaboration across Product, Engineering, and Business teams to deliver tangible business value.

Qualifications

• Experience designing GenAI applications in a business setting.
• Strong embedding models and vector search understanding.
• Proficient Python for data science and backend development.
• MLOps experience and deployment pipelines.

Responsibilities

• Design, develop, and deploy GenAI applications using Azure OpenAI, LangChain, and LangFuse.
• Build end-to-end AI solutions from backend services to user interfaces.
• Use vector databases (Weaviate) to manage embeddings and retrieval.
• Implement scalable MLOps pipelines for training, testing, and deployment.
• Develop and optimize retrieval-augmented generation (RAG) and NLP solutions.
• Apply traditional ML models to business problems and predictions.
• Collaborate with Product, Engineering, and Business teams to align AI with goals.
• Monitor and continuously improve model and system performance.
• Stay updated on GenAI developments and apply them where relevant.

Skills

GenAI/ML systems

Embedding models

Vector search

NLP fundamentals

MLOps

Azure Cloud

Python

Full-stack AI apps

Communication

Education

Bachelor's degree in quantitative field

Tools

Azure OpenAI Service

LangChain

LangFuse

Weaviate

Job description

Calibrate People in Sandton, Johannesburg, SA, is seeking a Data Scientist with deep GenAI/ML expertise to design, build, and deploy AI-powered applications at scale. You will work across backend and frontend layers, and implement MLOps to support reliable deployments.

The role requires strong knowledge of embeddings, vector search, NLP, and Python, with collaboration across Product, Engineering, and Business teams to deliver tangible business value.