AI​/ML Specialist - LLM Expertise

Fixtra is a consulting and delivery company specialized in enterprise technology and SAP ecosystems. We empower banks and insurers to move faster, think bigger, and innovate smarter. With SAP BTP, cloud-native solutions, and cutting‑edge AI, we help financial institutions break free from legacy constraints and lead in a world that never stops changing. We bring together bold ideas, deep SAP expertise, and a startup energy that never settles.

Whether it's automating core banking, migrating complex systems to the cloud, or building intelligent, customer‑first applications—we're already doing it, across Europe, the Americas, and Asia. Fixtra is growing fast—and so are our ambitions. We're not just building technology. We're building the future of financial services—secure, scalable, intelligent. Fixtra. Designed for bold thinkers. Built for fearless finance.

We are seeking a Senior AI/ML Specialist with deep expertise in Large Language Models (LLM) to shape the intelligence layer in our flagship Fixtra Wizard product. This is a high‑impact role where you will architect, implement, and optimize end‑to‑end AI pipelines—from prompt engineering to smart orchestration and seamless integration with modern apps and data layers. You’ll work hands‑on with the latest LLM APIs (OpenAI, Anthropic, open‑source LLMs), advanced vector databases (Pinecone, Weaviate, FAISS), and robust automation frameworks to deliver production‑ready AI features.

As both a technological innovator and trusted advisor, you will collaborate daily with product and engineering teams to drive best‑in‑class AI solutions for financial services. If you’re driven by the challenge of pushing new models into production, thrive in a fast prototyping and automation‑heavy culture, and love exploring new tools to elevate end‑user experience, we want to meet you.

• Design, build, and optimize AI/ML workflows focused on Large Language Models and NLP, ensuring robust, scalable, and maintainable production systems.
• Engineer prompt strategies and iteratively refine prompts to maximize the quality, reliability, and relevance of model outputs.
• Integrate LLMs (OpenAI, Anthropic, Gemini, open‑source) with app features—intelligent search, Q&A, summarization, and predictions—using frameworks like Lang Chain or custom orchestrations.
• Architect and implement vector‑based information retrieval with databases such as Pinecone, Weaviate, or FAISS.
• Develop, automate, and monitor end‑to‑end AI pipelines, leveraging containerization (Docker) and workflow orchestration tools (Apache Airflow, n8n, Prefect).
• Gather, preprocess, and clean large volumes of textual data from diverse sources, including web scraping and API integrations.
• Define and track success metrics for AI feature performance; run evaluations, A/B tests, and continuous monitoring in live environments.
• Champion AI ethics, safe content handling, prompt
  
  Ops, and responsible model usage throughout product development.
• Continuously explore, compare, and prototype with emerging LLMs, frameworks, and AI productivity tools to improve speed and quality of delivery.
• Collaborate with full‑stack and product teams to seamlessly embed AI‑powered features into user‑centric workflows.
• Mentor teammates in best practices for prompt engineering, model evaluation, and leveraging AI tools efficiently.
• Document solutions, discoveries, and lessons learned to build Fixtra’s internal knowledge base.

• 5+ years of hands‑on experience in AI/ML, NLP, or applied data science roles, with significant exposure to deploying LLM‑enabled applications.
• Proven expertise with LLM APIs (OpenAI GPT‑4, Anthropic Claude, Gemini, and/or open‑source LLMs) and advanced prompt engineering.
• Strong Python programming skills for data science, ML scripting, pipeline automation, and production readiness; familiarity with Type Script/JavaScript is a strong plus.
• Experienced with AI orchestration frameworks (Lang Chain, Llama Index, Haystack) for RAG and tool integration.
• Deep knowledge of vector databases (Pinecone, FAISS, Qdrant, Weaviate) for semantic search and embedding storage/retrieval.
• Skilled in Docker‑based model…