AI Founding Engineer

About Vivox

Vivox AI builds AI Agents for financial crime compliance teams in fintechs and banks. Our platform automates up to 90% of manual work, helping companies onboard more good customers, reduce operational costs and risks, and boost compliance efficiency by up to 80%.

Vivox’s AI Agents are trusted by global regulated financial institutions, and several customers have passed independent audits confirming Vivox’s compliance. The company is backed by leading European VC and C-suite executives from Barclays, Silicon Valley Bank, Apple, Onfido, and Google.

HQ is based in London, with the team distributed across the US and Europe. We’re a fast-growing team building the future of AI in compliance and financial crime prevention. If you’re excited about solving complex problems with real-world impact, join us and help shape the next generation of intelligent compliance systems.

The Role

We’re looking for an experienced AI engineer to take ownership of designing and implementing large-scale AI systems for document and text analysis.

You’ll build robust pipelines that extract, interpret, and structure information from unstructured sources (PDFs, HTML pages, Word files, registry APIs).

Your work will combine LLMs, RAG systems, agentic workflows (LangGraph), NLP, and Knowledge Graphs to turn raw data into reliable, explainable risk intelligence.

You’ll play a key role in designing multi-agent reasoning flows that connect document parsing, retrieval, and inference into a cohesive, production-ready system.

Core Responsibilities

• Design and implement document analysis pipelines capable of parsing, normalizing, and structuring data from heterogeneous sources (PDF, HTML, DOCX, JSON, registry APIs).
• Develop modular AI workflows that combine deterministic parsing (rules, regex, heuristics) with LLM/NLP reasoning for classification, extraction, and summarization.
• Design entity extraction and linking logic to unify information across heterogeneous documents and data sources.
• Build retrieval-augmented generation (RAG) systems to integrate external knowledge and improve model context using vector databases (OpenSearch, FAISS, Qdrant).
• Design and orchestrate multi-step reasoning and agentic workflows using LangGraph and LangChain — including tool-using, context routing, and decision logic between specialized sub-agents.
• Construct and maintain Knowledge Graphs representing entities, relationships, and risk indicators to enable reasoning and explainability.
• Implement evaluation and quality control pipelines — automatic benchmarks, validation sets, hallucination detection, and model drift monitoring.
• Develop model-serving layers for AI inference (FastAPI, Celery, asyncio) in collaboration with the platform team.
• Integrate and orchestrate third-party LLM APIs (OpenAI, Anthropic, Gemini, Mistral) through LiteLLM, LangChain, or LangGraph frameworks.

Our products are already in use, and your contributions will focus on enhancing existing features, scaling our platform, and developing new functionalities. Additionally, you'll have the opportunity to shape the company's future and share in its success through our stock options program.

We offer remote work but are open to supporting relocation to London or Barcelona in the future.

Key Responsibilities

• Implement and optimize LLM/NLP/NER workflows for applications like entity recognition, sentiment analysis, content filtering and categorization
• Build and optimize RAG systems.
• Ensure data transparency and traceability, implementing a system of references to justify conclusions.
• Collaborate with cross-functional teams to deliver impactful AI solutions

Our Stack

• Backend: Python (FastAPI, Celery, asyncio, Pydantic), PHP (Laravel)
• Storage: PostgreSQL, Redis, OpenSearch
• AI: LangChain, LiteLLM, LangFuse, OpenAI, GeminiAI
• Frontend: React, TypeScript, Tailwind
• Infra: AWS, Docker, GitHub Actions, Cloudflare

Core Requirements

• 4+ years of experience in AI/ML/NLP development, including production systems
• Strong Python engineering skills
• Practical experience with document parsing and information extraction
• Solid understanding of LLM pipelines, RAG, and vector databases
• Experience building agentic or multi-tool LLM systems (LangGraph, LangChain, LlamaIndex)
• Familiarity with Knowledge Graphs, embeddings, and data modeling for reasoning
• Strong analytical mindset and ability to design measurable evaluation pipelines

Nice-to-Have Skills

• Experience developing and deploying solutions using microservices architecture
• Hands-on experience with knowledge graphs and their integration into RAG systems
• Familiarity with task queue systems for workload management

What We Offer

• Competitive salary
• Professional development budget
• Relocation assistance to London or Barcelona
• Stock options with a transparent vesting schedule
• Access to the latest AI technologies and research
• Flexible, fully remote work environment
• Rapid career growth opportunities
• Innovation-Driven Culture: Minimal bureaucracy and a focus on solving challenging problems without unnecessary distractions