Ph3Company Overview /h3 pteam.blue is an ecosystem of successful brands working together across regions to provide customers with everything they need to succeed online. 60+ successful brands make up the group; with a team of more than 3000+ experts serving its 3.5 million customers across Europe and beyond. /p h3Position Overview /h3 pteam.blue is building the AI layer that runs across one of Europe's largest digital‑services ecosystems, powering hosting, domains, email, and SaaS for millions of SMBs. As Principal AI Solutions Engineer you will be the senior technical authority on AI systems end‑to‑end: from model research and fine‑tuning through agentic orchestration, real‑time inference, and production reliability. This is not a research‑only role and not an MLOps‑only role. You will do both, setting technical direction, shipping production AI, and raising the bar across a team that is moving fast. /p h3Key Responsibilities /h3 ul libAgentic AI Systems /b ul liArchitect and evolve our multi‑agent orchestration platform (currently built on Hermes / Multica), including plugin systems, tool‑use pipelines, observability hooks, and channel adapters (voice, telephony, messaging) /li liDesign and implement voice AI pipelines — STT (VibeVoice‑ASR, Whisper), real‑time TTS with streaming (VibeVoice‑Realtime), VAD (Silero), SIP/RTP telephony integration — with sub‑300 ms end‑to‑end latency targets /li liBuild and maintain RAG pipelines with retrieval quality measurement, re‑ranking, and hybrid search over vector + keyword indexes /li liDefine MCP server architecture and tool‑use contracts across internal and third‑party integrations /li /ul /li libModel Development Fine‑Tuning /b ul liFine‑tune and evaluate LLMs (LoRA, QLoRA, DPO) for domain‑specific tasks including customer support, classification, and structured extraction /li liEvaluate and benchmark model quality using automated evals, human preference data, and domain‑specific metrics (WER, DER, cpWER for speech; RAGAS / LLM‑as‑judge for RAG) /li liManage model lifecycle: experiment tracking, versioning, reproducibility, and promotion to production /li /ul /li libObservability Reliability /b ul liOwn the AI observability stack: Langfuse tracing, span‑level LLM call instrumentation, cost tracking, and quality regression alerting /li liDefine and enforce guardrails: hallucination detection, PII redaction, output safety scanning, and rate‑limiting across multi‑tenant deployments /li /ul /li libPlatform Pipelines /b ul liBuild data ingestion, preprocessing, and feature pipelines supporting model training and continual learning /li liDrive CI/CD for ML: automated eval gating, shadow deployments, canary releases, and rollback triggers /li /ul /li libTechnical Leadership /b ul liSet architectural standards for AI systems across the group; conduct design reviews and own ADRs for major decisions /li liMentor ML engineers and applied scientists; grow the team’s capabilities in production AI, not just prototype AI /li liCollaborate with Product and Commercial teams to translate business problems into ML problem formulations with clear success metrics /li liEngage with external research partners and track emerging work (arXiv, conference proceedings, open‑source releases) to identify signals worth productionizing /li /ul /li /ul h3Experience Skills /h3 ul li8+ years in ML Engineering, Applied AI, or Research Engineering with at least 2 years in a lead or staff‑level role /li liDeep, hands‑on experience with LLMs in production: fine‑tuning, RLHF/DPO, prompt engineering, RAG, and tool use /li liFluent in Python and the core ML stack: PyTorch, Transformers (HuggingFace), PEFT/LoRA /li liReal experience with LLM inference serving — vLLM, TensorRT‑LLM, or TGI — in a latency‑sensitive production environment /li liPractical knowledge of agentic frameworks: multi‑agent coordination, tool‑call orchestration, context/memory management, and observability (Langfuse, Opik, or equivalent) /li liExperience with speech AI (ASR/TTS pipelines) or real‑time audio systems is a strong plus /li liSolid understanding of MLOps: experiment tracking (MLflow/WB), model registries, containerization (Docker/Kubernetes), and CI/CD for ML /li liAwareness of LLM‑specific risk: hallucination, prompt injection, data leakage, fairness, and privacy — and how to mitigate them in production /li liStrong communication skills: you can write a crisp design doc, run a productive architecture review, and explain tradeoffs to a non‑technical stakeholder /li /ul h3Nice to have /h3 ul liExperience with voice pipelines end‑to‑end: VAD → ASR → LLM → TTS → SIP/RTP telephony /li liMulti‑hop RAG with self‑consistency, chain‑of‑thought reranking, or RAPTOR‑style hierarchical retrieval /li liFamiliarity with MCP (Model Context Protocol) server design and tool‑use contracts /li liContributions to open‑source ML projects or published work (arXiv, NeurIPS, ACL, Interspeech, etc.) /li liExperience with multimodal models (vision‑language, audio‑language) /li liKnowledge of quantization techniques (GPTQ, AWQ, GGUF) and their quality/latency tradeoffs /li /ul h3Right to Work /h3 pAt any stage, please be prepared to provide proof of eligibility to work in the country you’re applying for. Unfortunately, we are unable to support relocation packages or sponsorship visas. /p h3Diversity Inclusion /h3 pEveryone is welcome here. Diversity Inclusion are at our core. Far above any technical competence, we value respect, openness, and trusted collaboration. We do not tolerate intolerance. /p /p #J-18808-Ljbffr