AI Engineer
About SyntaAI
SyntaAI develops Syntasec, an AI-powered SAP security and GRC compliance platform deployed entirely within the customer’s own infrastructure, so SAP data never leaves their network. The platform combines deterministic security analysis with LLM-based reasoning to help enterprise SAP teams find, understand, and remediate security and compliance risk.
We are a small, founder-led engineering team. The product is live and running against real SAP systems.
Role overview
We are hiring an AI Engineer to build and improve the agent and reasoning layer of the platform. This is a hands-on engineering role. You will work directly with the CTO on production systems that run inside customer environments, where correctness and explainability matter more than novelty.
Key responsibilities
- Design, build, and improve LLM-based agents that analyse SAP security data and propose remediations for human approval
- Build and tune retrieval-augmented generation pipelines, including embedding strategy, chunking, and evaluation of retrieval quality
- Write and iterate on production prompts, and build evaluation harnesses to measure output quality against real data rather than intuition
- Develop and maintain backend services in Python and FastAPI, backed by PostgreSQL
- Work on the deterministic analysis layers that the AI reasons over, ensuring findings are grounded in evidence and never inferred from absent data
- Contribute to our MCP tool surface and integrations
- Participate in code review and maintain the engineering standards of a system that customers run inside their own networks
Requirements
- 3+ years of software engineering experience, with strong Python
- Hands-on experience building applications on top of LLM APIs, beyond prototypes and demos
- Practical experience with retrieval-augmented generation, embeddings, and vector search
- Solid grounding in backend engineering: REST APIs, relational databases, testing, and Docker
- The judgment to know when a problem should be solved deterministically rather than with a model
- Ability to work independently in an early-stage environment with limited process
Preferred:
- Experience with agentic systems, tool use, or the Model Context Protocol
- Experience with PostgreSQL and pgvector
- Exposure to enterprise software, security, compliance, or SAP
- Experience building evaluation frameworks for LLM outputs
You do not need prior SAP experience. We will teach you the domain. We cannot teach engineering judgment.
What we offer
- Direct collaboration with the founding team and real ownership of the AI layer of the product
- Work on production AI systems that run inside enterprise customer environments, not internal experiments
- Equity for full-time candidates
- Remote-first, with flexible working hours
Specific compensation details will be discussed with shortlisted candidates.
How to apply
Email [email protected] with your CV and a short note on something you have built with LLMs, what was hard about it, and what you would do differently now. Links to code are welcome.