Research Scientist, Full-time Intern

About Goaly At Goaly, our mission is to make custom AI affordable for every business. Our founding team comes from the front lines of top AI labs and tech giants (Meta MSL, TikTok AI, Google DeepMind, xAI, Microsoft Research, etc.), where we built large-scale training infrastructure powering trillion-parameter models and scaled GenAI models to a global user base. Now, we are building something we wish we had before: a platform that makes training and adapting custom AI affordable for all modern companies, not just Big Tech. Our north star is ambitious: for a domain-specific task, reach 90% of SOTA performance at less than 10% of the cost. To get a taste of what we are doing, see our first tech blog. Role overview As an AI Research Intern, you will be deeply involved in the frontier challenges of large-scale model training. You won\'t just follow instructions—you will personally research, design, build, and optimize systems that push technical limits. Your code and research results will integrate directly into our core product or be submitted to top academic conferences like NeurIPS, ICML, or CVPR. Core responsibilities Build next-gen architectures: Design and validate novel multimodal foundations (vision-language, audio-text) to surpass SOTA benchmarks. Own end-to-end training: Lead LLM development from data curation to inference optimization. Bridge research and production: Rapidly prototype top-conference methods (e.g., CVPR/ICLR papers) and deploy improvements to live systems—like reducing cross-modal alignment latency by 35%. Academic Publication: We provide full support (experimental resources + writing guidance) for breakthrough results, aiming for NeurIPS/ICML 2026. Requirements Ph.D. or Master\'s student in CS, AI, Math, or related fields (Exceptional undergraduates with strong research capabilities may be considered). LLM Combat Experience: Deep understanding of Transformer architecture, attention mechanisms, and Scaling Laws. Hands-on experience training or fine-tuning large-scale language models from scratch. Multimodal Background: In-depth research in cross-modal learning (e.g., CLIP, Diffusion Models, VQ-VAE, or Multimodal LLMs like LLaVA/Flamingo). Experience optimizing multimodal systems. Proficient in PyTorch or JAX. Clean coding style and strong command of Python. Adaptability: A fast learner with a strong sense of responsibility, capable of wearing multiple hats and handling cross-stack challenges. Bonus Points First-author publications at top conferences (NeurIPS, ICML, ICLR, CVPR, ACL). High-star open-source projects on Hugging Face or Gold/Silver medals in Kaggle competitions. Why Join Us? Expert Mentorship: Work side-by-side with experts who have led the R&D of top-tier AI models. Compute Freedom: Access to abundant GPU cluster resources—don't let your creativity be limited by compute. Flat Culture: Flat management structure that rejects office politics and values only technology and results. Full-time Opportunity: Outstanding interns will receive highly competitive full-time offers and early equity incentives. #J-18808-Ljbffr