About the job
The world’s most critical--and at risk--business applications have been neglected for far too long. Onapsis eliminates this blind spot by providing cybersecurity solutions dedicated to business-critical applications. Whether running on premises, in the cloud, or in a hybrid environment, Onapsis helps nearly 30% of the Forbes Global 100 understand the threats and risks across their SAP and Oracle landscapes.
We are seeking an AI Engineer to join our mission-driven team. In this role, you will drive the incorporation of Gen AI, ML, directly into our product design. You will be at the forefront of shifting our product strategy and we want you to champion next-generation intelligent systems that fundamentally elevate our cybersecurity solutions.
What you will be doing, your legacy:
You will be working directly with company Principal Engineers evaluating, scoping, proposing, and building features to fulfill business solution requirements to protect our customers. You will play a direct role in laying the technical foundation for a new product offering. Additionally, you will be working with Data Engineering and DevOps to deliver high-quality products and services while also working closely with security and IT professionals to ensure safe and secure best practices are followed.
Responsibilities:
Bridge Data & AI Systems : Partner with core data engineers to allow ingestion and consumption via Model Context Protocol (MCP)
Adapt & Tune Local Models Adapt open-weights Small Language Models (SLMs) using LoRA and QLoRA fine-tuning. Optimize and deploy these models to run securely on proprietary infrastructure containing sensitive data.
Architect & Implement Advanced Retrieval Systems: Design and implement hybrid retrieval architectures, including Retrieval-Augmented Generation (RAG), Cache-Augmented Generation (CAG), and Knowledge-Augmented Generation (KAG).
Leverage vector databases and structured knowledge graphs to deliver rich, grounded context to downstream models.
Implement production-grade LLM /NLI observability and evaluation tools to monitor latencies, token consumption, and agent behavior in real time.
Evaluate model outputs for factual grounding, accuracy, and compliance
Host, serve and scale fine-tuned open-weights models utilizing low-latency runtimes like vLLM
Translate complex agentic workflows and LLM concepts into clean, maintainable software architectures.
Qualifications:
5+ years of software, data, or systems engineering experience.
2+ years of hands-on experience building and deploying production-grade NLP, LLM, or multi-agent applications.
Strong proficiency in Python and REST API development.
Deep understanding of API architecture patterns and multi-tenant system design.
Experience working with distributed data stacks and platforms such as Apache Spark, Databricks or Snowflake.
Experience with a major cloud provider: AWS (preferred) / Azure / GCP.
Understanding of deep learning concepts and Transformer architectures.
Hands-on experience with model quantization formats
Experience with AI evaluation frameworks and LLM observability tools.
Experience with AWS Sagemaker / AWS Bedrock other AWS AI services is a plus.
Practical experience with LLM orchestration frameworks and multi-agent system design.
Familiarity with Model Context Protocol (MCP) specifications and vector database implementations.
What we offer:
A role in shaping the future of protecting the most critical applications that run the world's business and a career that grows as the company grows.
A unique culture of high achievement and teamwork.
Supportive and humble colleagues are the space's top problem solvers and innovators.
Financial security through competitive compensation and incentives (meal vouchers, private medical subscription, a generous leave package that includes 25 vacation days, 4 volunteering days through our OnaCares Program, work anniversary day off, a dedicated peer recognition platform, Focus Fridays, LinkedIn Learning).
Where & how we work
This Bucharest-based role offers a hybrid setup (1-2 days per week) at our One Park Cotroceni office, hired exclusively under a standard labor contract ( CIM ). Please note that we are only hiring direct employees and cannot support B2B collaborations for this position.
About Onapsis:
Onapsis protects the business applications that run the global economy. The Onapsis Platform delivers vulnerability management, change assurance, and continuous compliance for business applications from leading vendors such as SAP, Oracle, and others. The Onapsis Platform is powered by the Onapsis Research Labs, the team responsible for the discovery and mitigation of more than 1,000 zero-day vulnerabilities in business applications.
Onapsis is headquartered in Boston, MA, with egional offices in Heidelberg, Germany; Buenos Aires, Argentina; Texas, USA, and now in Bucharest, Romania; and proudly serves hundreds of the world’s leading brands, including close to 30% of the Forbes Global 100, six of the top 10 automotive companies, five of the top 10 chemical companies, four of the top 10 technology companies, and three of the top 10 oil and gas companies.
For more information, connect with Onapsis on LinkedIn or visit https://www.onapsis.com .
#LI-AC1
#LI-Hybrid