Senior AI Engineer 🇪🇺
Pennylane SAS29 days ago
Remote, Germany +4 moreSenior
Responsibilities
- Build and harden production agentic loops covering LLM orchestration, tool calling, streaming and events, error recovery, checkpointing, and resumability.
- Design context and memory systems, including retrieval, compaction, short- and long-term memory, and latency and cost trade-offs.
- Turn business capabilities into documented, versioned, tested tools exposed internally and through MCP, while steering agent behavior with prompts, skills, planning, tool-selection strategies, and guardrails.
- Implement security, permissions, audit trails, isolated execution, token budgets, and latency budgets for agentic systems.
- Develop evaluation practices using golden datasets, LLM-as-judge, human evaluation, A/B testing, regression tracking, quality metrics, and error analysis.
- Collaborate with product teams and accounting domain experts on ComptAssistant, Studio Assistant, MCP, document extraction, Autopilot bookkeeping, and revision use cases.
- Monitor emerging AI trends, state-of-the-art models, new LLM architectures, multimodal AI, and optimization techniques.
- Contribute to team roadmaps, cross-team projects, AI platform improvements, knowledge sharing, process design, and mentoring.
Requirements
- 5–8 years of experience and very strong Python skills.
- Hands-on experience building LLM and agentic systems at scale in production, including prompting, tool use, context construction, RAG, failure handling, state, and reliability.
- Experience treating evaluation as a first-class discipline, including golden datasets, LLM-as-judge, human evaluation, A/B testing, regression measurement, and edge-case analysis.
- Good understanding of applied LLM, machine learning, AI infrastructure, model serving, vector databases, cost, and latency.
- Strong technical, business, and product skills with the ability to communicate with non-technical domain experts.
- Fluency in English; French is not mandatory.
- Model fine-tuning and post-training experience with SFT, DPO, or RL is preferred.
- Experience with MCP and familiarity with Ruby are preferred.
Benefits
- 25 paid vacation days.
- Competitive compensation package and company shares.
- Home-office equipment budget and monthly coworking allowance.
- Access to Gymlib fitness spaces and wellness activities.
- Access to Busuu language learning.
- Latest Apple equipment.
- Remote work from the employee’s European country of residence, within a maximum two-hour time difference from CET, depending on team and role requirements.
- Regular company events, including Tech Days every two months and an annual company seminar.
- For France-based employees: French contract, 6–12 RTT, five weeks of PTO, Swile lunch credits, Alan Blue healthcare coverage, and local events.
Tech Stack
Categories