Lendbuzz is seeking a Software Engineer focused on Large Language Model (LLM) applications to help design, build, and optimize next-generation conversational agent technologies, aiming to provide underserved and overlooked borrowers with better access to credit.
Requirements
- Strong programming skills in Python, with experience in ML and data tooling (e.g., PyTorch, Pandas, NumPy, Scikit-learn)
- Experience with NLP techniques, LLMs, or machine learning fundamentals
- Preferred: 2+ years of professional software engineering experience, including scripting, data processing, or backend/ML pipelines
- Preferred: Experience deploying applications or models on cloud platforms, preferably AWS
- Bonus (not required): experience with real-time systems, WebSockets/streaming, RAG pipelines, vector databases, or ML evaluation frameworks, Genesys/Twilio
Responsibilities
- Software engineering for LLM-powered conversational agents, with an emphasis on practical implementation, reliability, and user experience
- Evaluate, fine-tune, and deploy LLM-based models and pipelines using REST APIs and internal microservices
- Implement prompt engineering, retrieval-augmented generation (RAG), tool-use pipelines, and conversation orchestration logic
- Investigate and integrate emerging technologies, particularly in real-time voice, streaming, and multi-modal interaction
- Analyze model outputs, user interactions, and system performance to drive iterative improvements
- Build and maintain high-quality datasets, including data cleaning, preprocessing, labeling workflows, and benchmarking for NLP tasks
- Own data quality, ensuring accuracy, reproducibility, and reliability across the data lifecycle
Other
- This is a hybrid position based in Boston, MA and requires 3 days onsite.
- Strong problem-solving ability and comfort working independently in a fast-moving environment
- We value independent and critical thinking.
- We’ve built a company and a culture around a resolute belief in the promise and power of diversity.
- We celebrate our differences, and are better when we have a variety of experiences, viewpoints, and backgrounds.
