Senior ML Engineer – LLM & Agentic Systems

About Us

At Novasign, we’re redefining the future of bioprocessing. Our platform, Novasign Studio, combines intelligent hybrid models, automation, and modern microservices (SOA) architecture to accelerate the development of life-saving therapies, next-generation enzymes, and sustainable food technologies.

We’re scaling fast with a growing customer base across biotech and life sciences. If you’re looking to make a real-world impact with cutting-edge machine learning and SaaS technology, we want to hear from you. We’re hiring ten new team members in the next four months, and this role is critical in driving our growth.

Role Summary

We are seeking a senior ML engineer with deep expertise in large language models, agentic systems, and advanced machine learning (including DNNs, RNNs, and modern architectures) to build context-aware AI assistants, knowledge extraction systems, and conversational interfaces for Novasign Studio. You will design and implement LLM-powered and agentic features that help bioprocess engineers interact naturally with our platform, extract insights from documentation, and accelerate their workflow through intelligent automation. This role focuses on productizing AI capabilities rather than fundamental research, requiring strong software engineering skills alongside deep AI/ML expertise.

Key Responsibilities

• Design and implement context-aware AI assistants and agentic systems integrated seamlessly into Novasign Studio’s user interface

• Build intelligent chatbots, agentic interfaces, and conversational systems that understand bioprocessing domain concepts and user workflows

• Develop knowledge extraction systems to process and understand technical documentation and user-generated content

• Design and implement provider-based LLM infrastructure supporting multiple model providers (OpenAI, Anthropic, local models)

• Create data collection pipelines (with proper user consent) for domain-specific fine-tuning of language models on bioprocessing knowledge

• Develop robust evaluation frameworks for AI system performance, safety, and domain accuracy

• Implement MLOps practices for LLM serving, monitoring, and automated performance tracking

• Fully understand and apply best practices like Modular Design, OOP, SOLID, DRY, KISS, Composition, and Inheritance

• Build and maintain RESTful and gRPC APIs for AI service integration with other platform components

• Collaborate with UX/UI teams to integrate AI features naturally into the platform workflow

• Collaborate with other ML engineers to address specific technical challenges and domain problems

• Ensure proper handling of data privacy, security, and user consent in AI systems

• Stay current with rapid developments in LLM capabilities and evaluate their applicability to bioprocessing workflows

Required Qualifications

• Master’s degree or higher in Computer Science, Machine Learning, AI, or a closely related technical field

• 5+ years of experience in software engineering, with at least 3 years focused on LLMs, NLP, and conversational AI systems

• Expert-level Python programming skills (core constructs, modules, packaging: UV, Poetry, pip)

• Deep expertise in modern LLM architectures (Transformers, GPT, Claude, etc.), deep neural networks (DNNs), recurrent neural networks (RNNs), and advanced ML architectures, including fine-tuning techniques

• Strong experience with LLM APIs (OpenAI, Anthropic, Hugging Face) and managing provider-based AI architectures

• Proficiency with Hugging Face ecosystem (Transformers, Datasets, Tokenizers), LangChain for agent orchestration, and conversation management

• Experience building production AI systems including model serving, monitoring, and performance optimization

• Experience with hyperparameter tuning and optimization frameworks (Ray Tune preferred, Optuna, or similar)

• Knowledge of GPU optimization and distributed computing for efficient model training and inference

• Strong development practices including FastAPI/gRPC for AI service APIs, Docker/Kubernetes deployment, and CI/CD workflows

• Knowledge of vector databases (Pinecone, Weaviate, Chroma) and RAG (Retrieval-Augmented Generation) architectures

• Experience with MLflow or similar tools for experiment tracking and model lifecycle management

• Understanding of AI safety, alignment, and responsible AI practices for production systems

• Strong grasp of key design patterns and practices (e.g., DDD, SOLID, DRY, KISS, Composition, Inheritance)

• Experience with OAuth2/OIDC protocols, JWT, and RBAC implementation for AI system integration

• Docker & Docker Compose, Ubuntu (WSL2) development environment

• Git workflows, CI/CD fundamentals, and Agile/Scrum collaboration

• Excellent written and verbal communication skills in English

Preferred Qualifications

We welcome applicants who meet most —but not necessarily all—of the preferred qualifications listed below.

Priority levels: ●● Highly Desirable | ● Desirable

• Experience with domain-specific fine-tuning of language models ●●

• React/TypeScript experience for AI feature frontend integration ●●

• Familiarity with other languages such as C#, Python, or Go for smoother integration ●

• Experience with Ray or distributed computing frameworks ●

• Model quantization and optimization for efficient serving ●●

• Knowledge of prompt engineering and few-shot learning techniques ●●

• Experience with evaluation frameworks for LLM safety and domain accuracy ●●

• Vector database optimization and similarity search algorithms ●

• Experience with conversational AI platforms and chatbot development ●●

• Model compression techniques (distillation, pruning, quantization) ●

• Experience with multi-modal AI systems (text, images, structured data) ●

• Knowledge of AI ethics, bias detection, and responsible AI practices ●

• Understanding of EU AI Act compliance requirements and implementation ●●

• Triton Inference Server or Seldon Core for model serving ●

• KServe or Seldon Core for model serving ●

• Kubeflow Pipelines or BentoML ●

• Knowledge of bioprocessing, chemical engineering, or scientific domains

• Working proficiency in German is a plus

What We Offer

• Innovation Culture: We are an international team. We value new ideas, open discussions, and constructive criticism. Your voice shapes our technological direction

• Professional Growth: Continuous learning opportunities and career development in cutting-edge software

• Meaningful Impact: Work on software that accelerates life-saving therapies, enzyme manufacturing and sustainable food production

• Competitive Package: We offer an attractive salary above industry standards, complemented by comprehensive benefits, including a free food allowance. In accordance with the IT collective agreement (minimum ST1 – Regelstufe), the minimum gross annual salary is €53,802; however, your actual compensation will reflect your skills, experience, and impact and will be significantly higher

• Full time (38,5 h/week) – 25 days of paid holidays per full calendar year

How to Apply

Please apply directly via our LinkedIn job posting (preferred). This route ensures your application is tracked correctly and reaches the hiring team the fastest.

If applying through LinkedIn is not possible, you may instead email us at [email protected] with the subject line “Senior ML Engineer – LLM & Agentic Systems” and include your CV, GitHub (or comparable) profile, and a short cover letter.

We review applications on a rolling basis and aim to respond within one week.