ML Engineer Offline

Format: Remote or Lviv / Full-time

🌟 We are looking for:
A strong ML Engineer with hands-on experience in the world of LLMs, open-source models, and detection systems. The ideal candidate is someone who’s “in the know” — quick to grasp research-based approaches, able to configure and scale large language models, and build detectors and datasets even with minimal input data.

🔧 Key Responsibilities:

• Work with LLMs (e.g., Huggingface, VLLM, TGI)
• Self-host and manage LLM infrastructure
• Optimize large-scale dataset processing
• Build and train detectors for LLM output / AI-generated content
• Work with open-source model architectures, including modification of final layers
• Prototype and evaluate model performance on test datasets
• Monitor key metrics such as latency, throughput, hallucination rate, accuracy, and reliability

🧠 What we expect:

• 2+ years of experience in LLMs, NLP, or AI research
• Knowledge of leading and up-to-date detection approaches based on academic publications
• Hands-on experience with Python, C, Huggingface, and self-hosting tools
• Familiarity with OpenAI and Anthropic prompt engineering guidelines
• Strong autonomy and ownership over technical solutions
• Excellent problem-solving skills and analytical thinking
• Experience designing and scaling trust & safety detection systems, including pipelines for data quality assessment and risk analysis

✅ Nice to have:

• Experience developing custom or fine-tuned LLMs
• Ability to gather and adapt datasets without relying on pre-built resources

💰 Compensation:
To be discussed individually, depending on experience and depth of expertise