Senior AI Engineer
Essentials
Job title: AI Engineer
Location: Bucharest ( near Piata Universitatii)
Type: CIM, hybrid
Technologies
TensorFlow or PyTorch
Docker, Kubernetes, MLOps
Offer
A key role in an exciting international expansion.
The opportunity to help build our office in Romania from the ground up.
Three days in-office (with remote work on Mondays and Fridays).
An opportunity to shape your career growth while contributing to the company’s success.
A dynamic position embracing “freedom under responsibility”.
If sales targets are met, all employees enjoy an annual destination trip.
Competitive salary and benefits.
Other location-specific benefits
25 vacation days
Client
Or client is the leader in AI-powered, cloud-based Software Asset Management. Their technology brings clarity to complex IT environments, helping users make smart, data-driven decisions and maximize software ROI.
Recognized with the Highest Growth Award and ranked #3 Overall Champion at the Main Software 50 Awards Nordics, they are scaling fast and looking for people who want to join the journey. They value experience, but even more, the person behind it.
Role
As AI Engineer your role will include to design, build, fine-tune, and deploy advanced AI models using state-of-the-art machine learning and deep learning techniques. This includes managing the entire lifecycle of LLM and NLP systems, from data preprocessing to production deployment, ensuring scalability, performance, and accuracy for real-world applications. Our client is currently building a new team in Romania, so you’ll have the opportunity to be among the first members to join it.
Responsibilities
You will develop and optimize LLMs and NLP models, using methods such as LoRA and adapters. You will also preprocess and structure large-scale datasets for training and evaluation. Furthermore you will also:
- Engineer prompts and prompts-chains for instruction-tuned models.
- Evaluate model performance using task-specific metrics (e.g., F1, ROUGE, BLEU, perplexity).
- Optimize training on GPU/TPU with mixed-precision and memory-aware strategies.
- Implement distributed training using frameworks like DeepSpeed or PyTorch DDP.
- Integrate RAG (retrieval-augmented generation) pipelines using vector databases such as FAISS or pgvector.
- Track experiments using MLflow, Weights & Biases, or similar tooling to ensure reproducibility and auditability
- Package and deploy models in production using Docker, CI/CD pipelines, and orchestration via Kubernetes.
- Leverage cloud services (AWS, Azure, GCP) for scalable training and inference.
Ensure data governance and privacy compliance (e.g., GDPR, HIPAA) across AI workflows
Requirements
Strong expertise in training and evaluating neural networks using TensorFlow or PyTorch.
Hands-on experience with deploying ML models in production using Docker, Kubernetes, and MLOps pipelines.
Capable of implementing supervised, unsupervised, and reinforcement learning systems and optimizing them for performance and accuracy.
Able to manage cloud-based training/inference environments and scalable APIs with robust CI/CD workflows.
In-depth knowledge of NLP systems including LLMs, transformers, and embedding-based models.
Experience with prompt engineering, LoRA, adapters, and RAG integrations.
Familiar with techniques for fine-tuning foundation models and building custom task-specific variants.
Familiarity with Agile methodologies and experience working in an Agile team.
Strong proficiency in written and spoken English
Apply today
If you meet the minimum requirements and are interested in applying for this position, please send your details to careers@key-talents.com with “AI Engineer”, in the subject line.