Hugging Face Jobs

I am looking for an experienced AI/ML developer to build a ChatGPT-like chatbot without using OpenAI or any external API. Requirements: Use an open-source LLM (such as LLaMA, Mistral, or similar). The model must be hosted on my own server (VPS or dedicated server). Web-based interface (chat UI). Ability to fine-tune or train on custom data (optional but preferred). Secure and optimized performance. Technical expectations: Python (FastAPI / Flask) or Node.js backend. Experience with Hugging Face / Transformers. Knowledge of GPU deployment (preferred). Please include: Previous AI/LLM projects. Recommended model and hardware requirements. Estimated server specs (RAM, GPU)

I’m building an early-stage B2B SaaS product (StrataOps) focused on workflow automation for property management companies. ​ For this MVP, I need a Python developer to build an AI-powered maintenance triage system that sits on top of a shared Gmail inbox. ​ The system should be able to: Monitor incoming maintenance emails from tenants and property managers in a shared Gmail inbox. ​ Classify each email into categories (e.g., HVAC, plumbing, electrical, general, etc.). ​ Detect urgency and basic priority based on rules and content. ​ Extract key structured fields (tenant details, property, problem type, severity, timing, etc.) into JSON. ​ Prepare draft replies for tenants, and suggestions for vendors or next actions. ​ The MVP will be delivered in two main phases: Core triag...

I need structured, beginner-friendly coaching that will take me from “hello world” to employable in Natural Language Processing. My goal is to understand core NLP concepts, build a small portfolio of projects, and feel confident discussing them in interviews. I already code a little in Python but have no formal NLP background. I’d like to cover the foundations—tokenisation, embeddings, transformers—then progress to hands-on mini-projects using libraries such as spaCy, Hugging Face, TensorFlow or PyTorch. Guidance on best practices for data preprocessing, model selection, evaluation and error analysis is essential, along with tips on how to present this work in a résumé or GitHub repo. Deliverables I have in mind: • A personalised learning ...

I’m evaluating security vulnerabilities in three Arabic-capable language models—Allam, Falcon, and Fanar—by running the Garak prompt-injection suite. My top priority is the technical implementation, with a particular emphasis on translating each of Garak’s 256 English attack prompts into clear, natural Arabic before the tests run. Here’s how the workflow looks: • Build a Python notebook that loads the three models from Hugging Face (PyTorch backend), pipes the Arabic prompts through Garak, captures logits and full responses, and writes everything to tidy CSV files. • Include bilingual testing so the notebook can toggle between the original English prompts and their Arabic counterparts, allowing side-by-side success-rate comparison. • P...