Data Scientist (Core AI / NLP Engineering)

Urgent requirement for Data Scientist (Core AI / NLP Engineering) is required for our client in Bahrain

 

Design and develop machine learning models to support AI-driven banking solutions is MUST

Python (PyTorch, TensorFlow, LangChain, Hugging Face, OpenAI API, Anthropic Claude, etc.) is MUST

LangGraph, AutoGen, CrewAI, Flowise, or similar agent frameworks is MUST

Prompt engineering and orchestration (LangChain, LlamaIndex, Semantic Kernel, DSPy) is MUST

 

Responsibilities

 

•         Design and develop machine learning models to support AI-driven banking solutions

•         Collaborate with data engineers to access and prepare data for modeling

•         Apply statistical and ML techniques to solve business problems (e.g., churn prediction, credit

•         scoring)

•         Evaluate model performance and optimize for accuracy, precision, and recall

•         Deploy models into production using MLOps frameworks and CI/CD pipelines

•         Ensure models are explainable, auditable, and compliant with regulatory standards

•         Work with business stakeholders to identify AI opportunities and define success metrics

•         Document model assumptions, data sources, and performance benchmarks

 

 Core AI / NLP Engineering

•         Python (PyTorch, TensorFlow, LangChain, Hugging Face, OpenAI API, Anthropic Claude, etc.)

•         LLM fine-tuning (LoRA, PEFT, prompt tuning)

•         Retrieval-Augmented Generation (RAG), vector databases (Pinecone, FAISS, Weaviate, Chroma)

•         Prompt engineering and orchestration (LangChain, LlamaIndex, Semantic Kernel, DSPy)

•         Knowledge of embeddings, tokenization, and transformer architecture

•         Cloud AI tools: AWS Bedrock, Azure OpenAI, Vertex AI, OpenSearch, ElasticSearch

•         Model evaluation: hallucination detection, grounding, and benchmarking (BLEU, ROUGE, TruthfulQA, etc.)

Software Engineering & Backend Integration

•         RESTful and GraphQL APIs, webhooks

•         Containerization and deployment (Docker, Kubernetes, CI/CD)

•         Authentication and user/session management

•         Data pipelines and microservices

•         Knowledge of frameworks like FastAPI, Flask, NestJS, or Express

•         Integration with enterprise data (SharePoint, Salesforce, SQL, internal APIs)

Agent Orchestration & Tooling

•         LangGraph, AutoGen, CrewAI, Flowise, or similar agent frameworks

•         Task-decomposition and reasoning chains

•         Function calling, tool use, and API chaining

•         Memory design (short-term vs long-term)

•         Context management and grounding strategies

Conversational UX / Design

•         Conversation design frameworks (Google CCAI, Microsoft Bot Framework, Voiceflow, Botpress)

•         Flow design and intent management (Dialogflow, Rasa, Cognigy)

•         Tone, empathy, and personality design for AI personas

•         A/B testing dialogue variants and measuring user satisfaction

Data & Infrastructure

•         Data pipelines (Airflow, dbt, Kafka)

•         Structured/unstructured data ingestion (PDFs, databases, APIs)

•         Feature store and model registry management (MLflow, Kubeflow)

•         Vector database deployment and optimization

•         Monitoring, logging, and drift detection

Governance, Security & Compliance

•         Model explainability (SHAP, LIME)

•         Bias/fairness audits and data privacy

•         Compliance with GDPR, ISO 42001, NIST AI RMF, and local banking regulations

•         Secure prompt logging and audit trails

Products & Strategy

•         Translating business problems into AI use cases

•         Roadmapping and budget planning

•         KPI design (accuracy, user satisfaction, automation ROI)

•         Vendor management (OpenAI, Anthropic, AWS, etc.)

•         Change management and user adoption