AI Research Scientist (LLMs)

At Lendbuzz, we believe financial opportunity should be more personalized and fair. We develop innovative technologies that provide underserved and overlooked borrowers with better access to credit. From our employees to our dealers, partners, and borrowers, we’ve built a company and a culture around a resolute belief in the promise and power of diversity. We value independent and critical thinking. We are seeking a highly skilled and motivated AI Research Scientist with a PhD in relevant field to lead our initiatives in conversational AI and agentic AI systems. In this role, you will drive the research direction, technical strategy, and execution of next-generation intelligent agents across the organization. You will work closely with engineering, product, and leadership teams to develop, evaluate, and deploy innovative AI capabilities that power real-time, high-impact user experiences.  This position is a hybrid role based in Boston, MA, requiring 3 days onsite. Key Responsibilities:

Research Leadership & Technical Direction

Lead research and development for conversational AI, agentic AI, and multi-step reasoning systems

Explore and prototype novel architectures, algorithms, and interaction paradigms across text, voice, and multi-modal interfaces

Define technical roadmaps, research priorities, and model evaluation strategies for agent behaviors, tool-use workflows, and context management

Develop agent frameworks including prompt architectures, retrieval systems (RAG), planning modules, and tool orchestration pipelines

Build robust real-time systems including voice streaming, low-latency inference, and continuous context tracking

Design, fine-tune, and deploy LLM-based and multi-modal models using internal infrastructure, REST APIs, and microservices

Run systematic experiments to evaluate model quality, reliability, reasoning, and user experience

Analyze conversational data, agent behavior traces, and performance metrics to guide iterative improvements

Develop benchmarking protocols, simulation environments, and evaluation tooling for agentic workflows

Build and curate high-quality datasets for training and evaluation, including preprocessing, labeling workflows, and data governance

Ensure data accuracy, reproducibility, and rigor across the research lifecycle

Collaborate with backend and ML infrastructure teams on scalable deployment, monitoring, and optimization of agentic systems

Partner with product teams to translate research into production features

Work with engineering teams to ensure reliability, safety, and performance of deployed models

Contribute to internal documentation, research reports, and knowledge-sharing sessions

Key Requirements:

PhD in Artificial Intelligence, Computer Science, Machine Learning, or a related technical field

Prior research experience with Machine Learning / AI required

Strong proficiency in Python, with experience in ML frameworks (e.g., PyTorch) and data tooling (NumPy, Pandas, Scikit-learn)

Demonstrated experience in NLP, LLMs, conversational AI, or machine learning research

Strong problem-solving skills and ability to operate independently while driving complex research initiatives

Experience designing experiments, evaluating models, and working with large datasets

Preferred Qualifications:

2+ years of research or applied ML experience in conversational agents, LLM systems, or real-time AI applications

Experience deploying models or research prototypes to cloud environments 

Expertise with RAG pipelines, vector databases, reasoning or planning modules, or agent tool-use systems