Data Scientist, NLP

Role Overview

As a Data Scientist focusing on AI at Candidly, you will work on designing, developing, and deploying AI-driven features that enhance our financial wellness platform. You will partner closely with the product and engineering teams to create solutions that leverage natural language processing (NLP) to optimize the user experience, improve engagement, and improve operational efficiencies. You will contribute to projects ranging from OCR to conversational AI and beyond, leveraging both structured and unstructured data to deliver innovative financial wellness tools.

Key Responsibilities

• LLM Product Development: Design and build machine learning models to support new product features, including AI-powered guidance, personalized product experience.
• Natural Language Processing: Develop and deploy NLP-based models to extract insights from user interactions, and create intelligent financial guidance.
• Text Classification & Sentiment Analysis: Develop models to classify user queries, tag content, and determine sentiment in user interactions to inform product enhancements.
• Named Entity Recognition (NER): Build and train models to extract and identify specific entities from text.
• Topic Modeling & Content Recommendation: Implement topic modeling techniques (e.g., LDA, BERTopic) to automatically categorize user inquiries and generate personalized content or advice.
• Intent Recognition: Design and deploy models to accurately identify user intents from text or voice input, enabling more intuitive user interactions and tailored advice delivery.
• Data Strategy & Engineering: Collaborate with data engineering teams to define data needs, build ETL pipelines, and deploy machine learning models in production environments.
• Model Evaluation & Tuning: Conduct rigorous testing, optimization, and continuous evaluation of deployed models to ensure they deliver impactful and reliable results.
• Research & Innovation: Stay on the cutting edge of AI/ML advancements and evaluate new techniques for application within the Candidly platform.
• Stakeholder Collaboration: Work closely with stakeholders across product, design, engineering, and leadership to translate complex AI concepts into actionable strategies and features.
• Data Storytelling: Present findings and insights in a compelling way to both technical and non-technical audiences, contributing to strategic decision-making.

Requirements

• Experience: 3-5 years of experience in a data science role, with a strong focus on NLP projects.
• Technical Skills:
• Strong programming skills in Python and proficiency with libraries such as TensorFlow, PyTorch, Scikit-Learn, NLTK, SpaCy, Transformers, and Hugging Face.
• Expertise in NLP techniques such as BERT, GPT, Seq2Seq models, and transformer architectures.
• Proficiency in designing and deploying NLP pipelines for named entity recognition, text classification, and text generation.
• Expertise in designing and deploying machine learning models for production.
• Familiarity with cloud platforms (AWS) and tools for model deployment and orchestration
• Experience with SQL and data visualization tools is a plus.
• A love of experimenting and working with new tools and technologies
• Soft Skills:
• Strong communication skills and ability to explain complex concepts to non-technical stakeholders.
• High adaptability and comfort working in a dynamic, fast-paced startup environment.
• Proactive, with a passion for building impactful solutions that address real-world challenges.

Preferred Qualifications

• Experience in FinTech, financial services, or consumer lending domains.
• Proven track record of developing AI/ML solutions focused on user engagement or personalized recommendations.
• Hands-on experience with conversational AI, reinforcement learning, or advanced NLP techniques.