Multimodal Research Engineer Position

Multimodal Model Developer Role Vegah LLC is currently seeking a highly driven individual to work on developing multimodal models and agentic applications. This role is designed for individuals who thrive in fast-moving environments, stay ahead of AI advancements, and can deliver real value with minimal guidance. Responsibilities include: Building, fine-tuning, and evaluating multimodal LLMs (text, image, video, audio). Designing autonomous AI agents & workflows using LangChain, AutoGen, CrewAI, LlamaIndex, Haystack, etc. Implementing retrieval-augmented generation (RAG) pipelines with vector databases (Pinecone, Weaviate, FAISS). Developing and deploying production-grade APIs (Flask/FastAPI, Docker, GitHub Actions, CI/CD). Conducting independent research on cutting-edge models, frameworks, and tools — then applying findings to real-world use cases. Participating in impromptu working sessions and meetings aligned to US CST hours. Requirements include: Strong foundations in Deep Learning & ML (CNN, RNN, LSTM, Transformers, Attention). Hands-on experience with PyTorch, HuggingFace, LangChain (or similar). Exposure to multimodal models (e.g., GPT-4o, Claude, Gemini, LLaVA, Kosmos-2). Excellent Python skills (NumPy, Pandas, PySpark). Experience with APIs, Flask/FastAPI, Docker, GitHub Actions. Familiarity with cloud (AWS/GCP/Azure) is a plus. Self-motivated, independent researcher who thrives on solving problems without step-by-step guidance. Ability to work US CST hours and be available for ad-hoc meetings.

Apply Now