我們使用Cookie和類似技術來改善您的體驗、分析網站使用情況並協助我們的營銷工作。我們收集的數據包括頁面訪問記錄和匿名使用統計。 私隱政策
職位預算: USD 400-600
發佈時間:
Job Title: Freelance/Contract AI Engineer — Local LLM Infrastructure Setup (Remote) Employment Type: Freelance / Contract, Part-time (~10–15 hrs/week after pilot) Job Function: Information Technology > Systems Administration / AI & Machine Learning Location: Remote (client based in Central, Hong Kong) Salary: HK$400–900/hour, commensurate with experience (adjust before posting) Job Description: A private individual client in Hong Kong is looking to engage a freelance AI/infrastructure engineer to build a home-office local AI lab. The engagement is fully remote — no site visit required — with secure remote access arranged at the start of the project. Responsibilities: Configure and harden a 128GB unified-memory Linux laptop for local large-language-model inference Deploy and benchmark local LLMs (e.g. Gemma, Qwen-VL, GLM-Air) via Ollama/LM Studio/vLLM, served through a standard API Build a Python automation pipeline that ingests financial statement documents (PDF/email/cloud storage), extracts structured data using a local vision-language model, and updates a master spreadsheet, with a human-review checkpoint for low-confidence extractions Set up a multi-model research-assistant workflow combining local and cloud AI models Establish a versioned, portable notes/knowledge-base system shared across tools Package the environment for portability (containerized services, encrypted backups) Optional, only if experienced: read-only integration with the Interactive Brokers API for portfolio monitoring/reporting — no trading execution or decision-making software of any kind Provide short 1-on-1 training sessions and plain-English documentation so the client can operate the system independently after handover Follow a strict security protocol: no client credentials or real data are used until formal handover; all engineer access is revoked at project close under a documented process Requirements: Strong professional experience with Linux system administration, Docker, and Python Hands-on experience deploying and serving local large language models (Ollama, vLLM, or LM Studio) and working with quantized model formats Comfortable delivering an entire engagement remotely via screen-share and remote-access software Fluent written and spoken English Advantageous: experience with the Interactive Brokers API, AI agent orchestration frameworks, and unified-memory AI hardware platforms (e.g. AMD Ryzen AI Max, Apple Silicon) Other terms: NDA required. Paid pilot task precedes the ongoing engagement.