Mô tả công việc
AI Agent & Chatbot Development
Ensure chatbot responses are accurate, contextual, and aligned with public health values.
Design and implement AI- driven chatbots using LLMs (e.g., GPT, LLaMA, Mistral, etc.).
Build domain- specific pipelines to integrate organizational/public health knowledge into conversational agents.
Search & Personalization
Develop semantic search engines and retrieval- augmented generation (RAG) pipelines.
Implement personalization features (user profiling, adaptive responses).
Optimize relevance and ranking for search results using ML/AI techniques.
Virtual Assistant Capabilities
Extend AI agents to support virtual assistant features: task reminders, workflow automation, and knowledge support.
Integrate with productivity tools (calendars, CRM, digital platforms).
Technical Development & Integration
Ensure scalability, performance optimization, and responsible use of AI.
Fine- tune and deploy LLM models in secure and efficient environments.
Collaborate with cross- functional teams (Product Owner, developers, domain experts, QA).
Data Engineering & Preparation
Collect, clean, and preprocess structured and unstructured datasets for model training.
Build and maintain pipelines for knowledge ingestion (documents, APIs, databases).
Ensure data quality, labeling consistency, and bias mitigation.
Model Training, Evaluation & Deployment
Apply evaluation metrics (accuracy, precision, recall, F1, BLEU/ROUGE for NLP).
Train, fine- tune, and evaluate ML/LLM models on custom datasets.
Automate deployment workflows using CI/CD pipelines and MLOps best practices.
Monitoring & Continuous Improvement
Implement feedback loops for continuous learning and model updates.
Detect and mitigate model drift or degraded performance.
Monitor model performance post- deployment (latency, accuracy, hallucination rates, user adoption).
Security, Ethics & Compliance
Ensure compliance with data protection and ethical AI principles (privacy, fairness, transparency).
Implement safeguards to reduce risks of hallucinations, misinformation, and bias in chatbot outputs.
Innovation & Documentation
Stay updated with AI/ML advancements, particularly in LLMs, agent frameworks, and healthcare AI.
Document development processes, model architectures, and APIs.
Contribute to knowledge- sharing sessions and training for internal teams.
