The role
We are hiring our first machine-learning specialist to own the full lifecycle of training and fine-tuning large language models used by Veris AI customers. You will split your time between applied research (new fine-tuning methods, dynamic benchmarks) and production engineering (scalable training pipelines, robust deployments).
What you will do
- Evaluate and compare open-source LLMs for new tasks.
- Fine-tune models with SFT, RLHF, and reinforcement fine-tuning (RFT).
- Design and implement training workflows in PyTorch/Transformers with efficient CUDA kernels.
- Build and maintain the code that moves research ideas into production services.
- Reduce latency, memory, and cost of model inference.
- Create dynamic benchmarks that track real-world agent performance.
- Publish technical results, write internal docs, and present at conferences.
What you bring
- MSc or PhD in CS, EE, Statistics, or a related field.
- Peer-reviewed publications in machine learning or NLP.
- Hands-on experience training or fine-tuning LLMs at scale.
- Strong software engineering skills (Python, PyTorch, CI/CD, cloud GPUs).
- Ability to work independently, set your own research agenda, and communicate findings clearly.
Nice to have
- Prior work on reinforcement learning with human or programmatic feedback.
- Contributions to open-source ML frameworks.
- Experience building internal tools for dataset management, experiment tracking, or evaluation dashboards.
Why Veris
You will join a small team that ships quickly and cares about scientific depth. Your work will directly shape how enterprises train and deploy autonomous agents in high-stakes settings. Competitive salary, meaningful equity, top-tier health benefits, and the compute you need to get the job done.