AI Engineering

This is a remote position.

The Castillian was placed successfully with one of the clients in the Telecommunication Sector and is currently focused on different responsibilities as follows:

• Working on a new initiative to develop documentation processes, create prompting guidelines, and establish a company's AI policy accessible to the entire team.
• Working on multiple AI initiatives, mainly on Chatbots, the Document Control project, Orchestrator, and other AI chatbot solutions.
• Developing and refining AI-driven workflows for the Document Control project to automate repetitive tasks and improve internal team efficiency, particularly for the Finance team, while attending final adjustments and reviews.
• Designing, enhancing, and maintaining AI chatbots by updating AI behavior, implementing multilingual capabilities, expanding product offerings, and refining intelligent behavior, reopening tasks based on new requirements, and supporting additional automated testing and UAT cycles.
• Collaborating closely with the team, participating in requirement-finalization meetings with stakeholders, and implementing agreed changes before production deployment.
• Developing and testing intelligent AI to AI communication, allowing information to be transferred or escalated between different AI chatbots.
• Identifying and automating internal operational pipelines using n8n to improve employee efficiency and reduce manual effort.
• Delegating tasks to team members, reviewing their work, and ensuring quality and alignment with project goals.
• Leveraging large language models (LLMs), including text generation, sentiment analysis, and language understanding.
• Fine-tuning and transfer learning pre-trained LLMs.
• Integrating LLM-based models into applications and platforms by ensuring seamless interaction with existing workflows and data pipelines.
• Preparing and preprocessing text data for training LLMs, including cleaning, tokenization, and handling of various data sources (including APIs, databases, unstructured data).
• Continuously monitoring the performance of the LLM-based models to ensure they meet the defined metrics and objectives. This includes tracking metrics such as accuracy, response time, and user satisfaction scores for chatbots or sentiment analysis models.
• Monitoring security-related events or vulnerabilities in the LLM-based models or infrastructure supporting them.
• Implementing security monitoring tools and practices to detect and mitigate risks such as unauthorized access, data breaches, or model poisoning attacks.