Agentic Systems Engineer
AI Tools
Tech Stack
Agent Workflow
Building an agentic platform that continuously searches, analyses, prototypes and reports. Deploying agents for literature scanning, RAG pipelines, code review integration, and automated experimentation with guardrails and evaluation frameworks for LLM-based agents.
About the Role
Shen.AI delivers digital, smartphone-accessible health & wellness diagnostic tools that run locally on your mobile device. Based on a 30-second video scan of the human face, they extract over 25 health markers such as Heart Rate, HRV, Blood Pressure, Stress Level and more.
They are building an agentic platform that continuously searches, analyses, prototypes and reports — so their scientists and engineers can focus on breakthroughs.
As an Agentic Systems Engineer you will design, implement and operate this platform, giving Shen.AI a 10x faster innovation cycle.
Key Responsibilities:
- Deploy agents to scan scientific literature and patents, summarizing findings
- Build data pipelines ingesting knowledge from Slack, Confluence, JIRA, and emails; implement retrieval-augmented generation
- Integrate agents with GitHub for code review, log analysis, and problem-solving
- Create sandboxed environments for algorithm prototyping and optimization
- Orchestrate automated experiments with reproducible metrics
- Develop guardrails and evaluation frameworks for LLM-based agents
- Generate weekly executive summaries for technical and leadership teams
- Analyze biosignal data for pattern identification
- Ensure privacy-by-design and secure access to data
Required Qualifications:
- MSc / PhD in Computer Science, AI, Mathematics, or related field
- 4+ years of experience in software engineering or ML, including production use of LLM frameworks (LangChain, LangGraph, CrewAI or similar)
- Strong proficiency in Python and related libraries; ability to develop complex, testable workflows
- Hands-on experience with GCP (Cloud Storage, Vertex AI, Cloud Run/K8s), GitHub, CI/CD, Docker/Kubernetes
- Knowledge of RAG, vector databases (Pinecone, Weaviate, pgvector) and ingestion pipelines for unstructured data
- Ability to design evaluation pipelines and manage LLM cost/latency trade-offs
Nice-to-Have:
- DSP/computer vision experience with biomedical signal processing
- Slack, Confluence, JIRA, Sentry, GitHub integrations
- C++/CUDA optimization knowledge
Benefits:
- Unlimited paid vacation
- Private healthcare
- Remote/on-site flexibility (Wroclaw, Poland or Tallinn, Estonia)
- Competitive salary
- Equipment choice
- Collaborative team culture