AI Engineer

Our client is an international product company developing AI-powered digital products. The team works on Generative AI solutions combining conversational AI, image generation, and video generation.

We are looking for a GenAI Engineer / Python Developer to join the AI team and work on the model layer, building and improving production-grade generative AI pipelines.

What you will do

• Own and develop parts of the chat pipeline, including prompt construction, character personality conditioning, short-term memory, follow-up context, and response post-processing.
• Integrate and evaluate new LLM, image, and video generation models, comparing them in terms of quality, cost, and latency.
• Train and retrain LoRAs on curated datasets to maintain character consistency across image and video generation.
• Build and improve video generation pipelines, including multi-step scenarios, clip sequencing and reuse, duration control, and automated output validation.
• Build agentic workflows and automations for content production at scale.
• Run structured prompt experiments using datasets, prompt versions, and evaluations.
• Work with Langfuse for tracing and evaluation.
• Maintain the content policy layer, including topic restrictions, banned-word handling, and policy validation.
• Work with GPU inference infrastructure such as RunPod and similar providers.
• Optimize AI systems for cost, latency, and reliability.
• Monitor model and generation metrics and troubleshoot production issues.
• Work in two-week sprints together with backend, product, and QA teams.

What we are looking for

Must have

• 3+ years of professional Python development experience, including production backend services.
• Experience with async patterns, queues, and microservices.
• Hands-on experience integrating LLMs or Generative AI models into a real production product.
• Strong understanding of prompt engineering, structured outputs, fallback routing, and model failure handling.
• Practical experience with diffusion models.
• Experience with LoRA training / fine-tuning, checkpoint evaluation, or image-to-video generation.
• Experience building multi-step generation or agentic workflows.
• Experience with Docker, CI/CD, and GPU inference providers such as RunPod, fal.ai, or similar.
• Experience with AWS and MongoDB.
• Strong production engineering skills, including monitoring, debugging, and troubleshooting live systems.
• Good written and spoken English.
• Comfortable working on an 18+ product.

Nice to have

• Experience with Langfuse — datasets, experiments, and tracing.
• Experience with video generation models, such as Wan or similar.
• ComfyUI workflows.
• Self-hosted model serving using vLLM / TGI.
• Experience with conversational AI memory systems, including vector retrieval and session memory.
• Experience with content moderation or safety classifiers.

What we offer

• The opportunity to work on a modern Generative AI product.
• Challenging technical tasks at the intersection of Python, LLMs, image/video generation, and AI infrastructure.
• The opportunity to influence the development and architecture of AI solutions.
• An international and collaborative team.
• A dynamic product environment where you can see the impact of your work quickly.

If you have experience not only integrating AI APIs but actually building and shipping GenAI solutions in production, we would be happy to hear from you!

Відгукнутись на вакансію