Project Title:
Higher Education Acceleration and Transformation Project (HEAT)

Sub Project Title:
AIMScribe: AI-Driven Scribing System with Prescription Counseling Tools to Streamline Clinical Documentation, Enhance Medication Safety, and Patient Empowerment in Bangladesh.

Position Type:
Full-time, Contractual (for the duration of the project)

About the Role:
We are looking for an LLM Engineer with strong NLP and software development skills to work on fine-tuning and deploying large language models. This role combines research with product development, contributing from prototype to production and helping build NLP infrastructure for real-world applications.

Key Responsibilities:

• Design, implement, and optimize Speech to Text models for Transcribing Doctor-Patient consultation audio in real time.
• Develop and maintain robust text data generation and processing pipelines.
• Fine-tune and evaluate large-scale pre-trained LLM models for domain-specific tasks.
• Build robust NLP pipelines incorporating Automated Speech Recognition (ASR), Speaker Diarization, Named Entity Recognition (NER), and Text Summarization model from research to deployment.
• Stay up to date with the latest advancements in LLMs (RAG, MCP, tool use, agentic systems) and apply them effectively in projects.
• Integrate NLP components into larger production systems, ensuring scalability and performance.
• Follow best practices in software engineering, version control, testing, and cloud-based deployment.

Qualifications and Skills:

• Bachelor’s or Master’s in Computer Science, Data Science, or a related field.
• 1+ years of professional experience in LLM and Natural Language Processing.
• Strong proficiency in Python, especially for data processing and model development.
• Solid understanding of NLP concepts such as tokenization, embedding, language modeling, transformers, and sequence modeling.
• Demonstrated expertise in training, fine-tuning, and optimizing STT and advanced speech-to-speech foundation models.
• Strong background in signal processing, acoustic modeling, speech enhancement, and noise reduction techniques.
• Experience with LLM fine-tuning, prompt engineering, pretraining, and downstream evaluation.
• Hands-on experience in building data pipelines for training and evaluation.
• Experience working on end-to-end NLP product development (research to deployment).
• Deploy models and pipelines via FastAPI and AWS, and monitor/debug issues in production
• Exposure to MLOps practices, model deployment, and monitoring in production environments.
• Deep knowledge or hands-on experience with agentic LLM systems, tool calling, retrieval-augmented generation (RAG), and Model Context Protocol (MCP) workflows.