AI Research Internship

About Lexsi Labs

Lexsi Labs is a frontier lab focused on building aligned, interpretable, and safe Superintelligence. Their mission is to create AI tools that empower researchers, engineers, and organizations while maintaining transparency and safety.

About the Role

This is a full-time AI Research Internship for 6 months, with the potential for extension or a full-time offer. Interns will work on large-scale industry problems and advance AI technologies.

Key Responsibilities

Collaborate with research and engineering teams on one of the following areas:

• Library Development: Architect and enhance open-source Python tooling for alignment, explainability, model alignment, uncertainty quantification, robustness, and machine unlearning.
• Explainability & Trust: Improve and find new observations using XAI techniques (DLB, LRP, SHAP, Grad-CAM, Backtrace) across text, image, and tabular modalities.
• Mechanistic Interpretability: Probe internal model representations and circuits using activation patching, feature visualization, and related methods.
• Uncertainty & Risk: Develop, implement, and benchmark uncertainty estimation methods (Bayesian approaches, ensembles, test-time augmentation) and robustness metrics for foundation models.
• Tabular Foundational Models (Orion): Improve and launch new tabular foundational model architectures and work on the open-source library TabTune.
• Reinforcement Learning: Explore new ideas and algorithms around RL and a new RL fine-tuning library.
• Research Contributions: Author and maintain experiment code, run systematic studies, and co-author whitepapers or conference submissions.

Requirements & Tech Stack

• Strong Python expertise for writing clean, modular, and testable code.
• Deep understanding of machine learning and deep learning principles with hands-on experience with PyTorch.
• Comprehensive knowledge of Transformer architectures and fundamentals, including attention mechanisms, positional encodings, tokenization, and training objectives in models like BERT, GPT, LLaMA, T5, MOE, Mamba.
• Familiarity with Git workflows, packaging, documentation, and collaborative development practices.
• Excellent communication, peer code reviews, and agile teamwork skills.

Preferred Domain Expertise (Any one is good)

• Applied experience with XAI methods such as DLB, SHAP, LIME, IG, LRP, DL-Bactrace, or Grad-CAM.
• Familiarity with mechanistic interpretability techniques like circuit analysis, activation patching, and feature visualization.
• Hands-on experience with uncertainty estimation techniques like Bayesian methods, ensembles, or test-time augmentation.
• Experience with model quantization and pruning for optimization.
• Experience with LLM Alignment techniques, including prompt crafting, reward modeling, and human-in-the-loop fine-tuning.
• Experience with Tabular Foundational Models (e.g., Orion, TabPFN, TabICL).
• Practical work with post-training adaptation and fine-tuning methods (e.g., LoRA, adapters, instruction tuning, knowledge distillation).

Additional Experience (Nice-to-Have)

• Publications in relevant conferences (e.g., CVPR, ICLR, ICML, KDD, WWW, WACV, NeurIPS, ACL, NAACL, EMNLP, IJCAI) or equivalent research experience.
• Prior contributions to AI/ML libraries or tooling.
• Exposure to risk-sensitive applications in finance, healthcare, or similar fields.
• Familiarity with large-scale training infrastructures for performance optimization.

What We Offer

• Opportunity to address high-stakes AI challenges in regulated industries.
• Access to compute resources, including GPUs, cloud credits, and proprietary models.
• Competitive stipend with potential for full-time conversion.
• Authorship opportunities on papers, technical reports, and conference submissions.