Responsibilities- Build agentic capabilities for TeamViewer Tia and the Agentic Ecosystem, including orchestration, tool use, retrieval, memory, and multi-step task execution.
- Own evaluation discipline by defining quality criteria, building datasets and harnesses, and using evaluation results to guide release decisions.
- Engineer model context through retrieval strategy, chunking, ranking, caching, compaction, and prompt assembly.
- Improve quality through systematic iteration on prompts, tool design, model selection, and orchestration patterns.
- Take AI capabilities to production and manage latency, cost, rate limits, fallback behavior, and quality drift.
- Partner with product management, security, and platform teams to deliver secure, reliable AI-enabled features.
- Apply fine-tuning or smaller specialized models where evaluation evidence supports it.
- Work agent-first in your own engineering and help keep team context, tooling, and delivery patterns in strong shape.
Requirements- 8+ years of industry experience with strong Python expertise, solid software engineering fundamentals, and a proven track record of delivering production systems.
- Hands-on experience building model-based and agentic systems in production environments, including tool calling, structured outputs, retrieval, orchestration, and multi-agent architectures.
- Proven ability to evaluate AI systems responsibly by defining quality criteria, building evaluation frameworks, interpreting results, and identifying real-world failure modes that evaluations may not capture.
- A data-driven approach to prompt and context engineering, using retrieval strategies, instructions, tooling, and model context to deliver measurable improvements.
- Strong understanding of the current AI model landscape and the ability to make pragmatic decisions across providers, architectures, cost, latency, reliability, and quality requirements.
- Working knowledge of MCP or comparable tool protocols, agentic development environments, and the security implications of providing models with access to tools and enterprise systems.
- Regular use of AI coding agents, combined with critical review practices and accountability for the correctness, security, and maintainability of delivered software.
- Deep understanding of common AI failure modes, including hallucinations, context degradation, prompt injection, non-determinism, and silent regressions, along with effective mitigation strategies.
- Strong communication skills, the ability to engage effectively with both technical and non-technical stakeholders, and a collaborative mindset for working across EMEA teams while helping shape high-quality agentic engineering practices.
What we offer- Competitive compensation and bonuses
- Flexible PTO and paid holidays
- 401(k) with employer matching
- Comprehensive Health insurance package including 100% employer-paid medical coverage
- Up to 12 weeks of Parental Leave
- Basic Life Insurance, Short-Term & Long-Term Disability, 100% employer-paid
- Quarterly teambuilding events, leadership luncheons, and companywide "All Hands" meetings
- Open door policy and business casual dress code
- We celebrate diversity as one of our core values. Join c-a-r-e and lead change initiatives together with us!
Work location for this position is Austin, TX.
Department Research & Development Locations Austin Remote status Hybrid Employment type Full-time Type of Job Non Student
