Software Engineering Intern - LLM
Role Description
Beno Technologies invites passionate and talented individuals to join our dynamic internship program focused on advancing our AI-driven chat agent through experimentation and innovation. In our engineering culture, exploration drives progress, empowering us to push the boundaries of conversational AI. Your contributions will directly influence the evolution of our chat agent, helping the Beno Engineering team evaluate cutting-edge models, integrate real-time voice capabilities, and enhance user experiences.
This year’s project builds on our prior work: in 2022, we developed a chat agent using OpenAI’s ChatGPT, and in 2023, we evaluated local LLMs to match its performance. Now, we aim to explore newer models like DeepSeek and investigate real-time voice conversation features. Your role will involve benchmarking these models, fine-tuning responses, prototyping voice-enabled interactions, and delivering actionable insights to elevate our product’s competitiveness.
Who We're Looking For
We seek candidates currently enrolled in a postgraduate program in computer science, computational linguistics, data science, or a related field. Ideal candidates thrive on solving complex technical challenges, enjoy hands-on experimentation, and have led projects in AI or NLP during academic or professional work.
A strong foundation in Natural Language Processing (NLP), Large Language Models (LLMs), or speech processing is essential. Experience with LLM evaluation, fine-tuning, or prompt engineering—such as familiarity with OpenAI APIs, Hugging Face, or similar frameworks—will be critical. Proficiency in Python and libraries like PyTorch or TensorFlow is required, and knowledge of voice streaming APIs like WebSocket or speech-to-text tools like Whisper is a plus. The ability to prototype conversational interfaces, whether web-based, CLI, or voice-enabled apps, will round out the technical profile.
Beyond technical expertise, we value curiosity about emerging LLMs such as DeepSeek, Claude, or Mistral, as well as voice AI trends. Candidates should be self-driven, with a structured approach to testing hypotheses and documenting results. Excellent communication skills are vital for presenting findings and collaborating with cross-functional teams.
Benefits
Our robust mentoring approach has significantly benefited past interns, paving the way for them to secure full-time positions, including roles within our engineering team. Former interns consistently recommend our program to their peers, with many showcasing their experiences on LinkedIn.
While we do not provide compensation for short internships, outstanding performance during an extended internship may lead to compensation after internal approvals. Recent internship projects have involved contributing to open-source initiatives, providing candidates with tangible and referable contributions that can significantly enhance their careers.
Our internship will be fully remote; in rare cases, we might require field visits for some studies. This is rare and highly dependent on the nature of the project. For example, a project might require visiting our car rental warehouse to conduct studies or collect images of cars as input to our study. In case the university requires in-person attendance, we can accommodate this (subject to availability).
Duration
The internship typically spans 5–6 weeks, though we remain open to extensions for ambitious projects. By the conclusion, you’ll deliver a comprehensive report comparing LLM performance, a prototype of a voice-enabled chat interface (if applicable), and a final presentation to the engineering team.