Data Scientist (Junior) — CodeRound AI / Healthcare AI Startup

About the Role & Client Company

Key Responsibilities

Model Engineering & LLM Customization

• NLP & ML Development: Develop, implement, and maintain Python-based NLP and Machine Learning models tailored for healthcare applications and clinical documentation.
• LLM Adaptation: Work with Large Language Models (LLMs) and transformer-based architectures to build, adapt, and customize AI solutions for clinical coding and medical tech use cases.
• Model Evaluation & Optimization: Evaluate state-of-the-art open-source and proprietary models; fine-tune hyperparameters and prompt architectures to maximize accuracy, reliability, and inference performance under operational constraints.

Execution & Stakeholder Collaboration

• Requirement Translation: Translate business objectives and clinical workflow requirements into practical ML/NLP solution tasks with full ownership of execution deliverables.
• Domain Customization: Analyze performance metrics on existing models and customize architectures to handle specialized clinical sub-domains and client-specific electronic health record (EHR) requirements.
• Technical Communication: Articulate model outputs, technical trade-offs, confidence scores, and limitations clearly to both engineering leads and non-technical stakeholders.

Candidate Qualifications & Requirements

Required Experience & Core Skills

• Experience: 2+ years of hands-on professional experience building and deploying Python-based Machine Learning, NLP, or Applied AI solutions.
• Programming Mastery: High proficiency in Python and its core machine learning ecosystem.
• NLP Foundations: Strong understanding of core NLP techniques, including text preprocessing,tokenization, vector embeddings, entity extraction, semantic search, and classification.
• LLM Expertise: Practical understanding of transformer architectures (e.g., BERT, LLaMA, Mistral, GPT variants), LLM fine-tuning techniques (LoRA/QLoRA), embeddings, and real-world AI limitations.
• Model Benchmarking: Proven ability to evaluate, fine-tune, and optimise ML/NLP pipelines for precision, recall, and overall model reliability.

Key Cultural & Behavioural Fit

• Ownership Mindset: Ability to take direct ownership of feature tasks from day one within a fast-paced Series A startup environment.
• Problem-Solving Agility: Eagerness to rapidly iterate on state-of-the-art research models and adapt them to noisy, complex clinical datasets.

Role Summary