AI / Software Engineer (Full-Stack & Systems)

About FactTrace FactTrace comes from two roots: Fact (verified truth) and Trace (the ability to follow and verify). Together, they express a vital idea: truth that can be followed, even as its form changes. We are a fast-moving, high-intensity startup building the truth infrastructure for the AI era—a foundation that preserves the integrity of meaning as language moves between people, systems, and machines. Working alongside top talent from the Cavendish Laboratory at the University of Cambridge , we are solving the next generation of AI problems. By 2030, deterministic verification of textual information will not just be an advantage; it will be a absolute requirement. FactTrace exists to build that layer. Located in the heart of Cambridge's tech ecosystem, we offer a hyper-collaborative environment where technical rigor, extreme focus, and rapid execution drive everything we do. The Mindset We are looking for early-career, highly committed Engineers who want to shape the future of AI. Building foundational infrastructure requires generalists with an intense drive, laser focus, and a commitment to company-wide success. We are not looking for engineers who only want to tweak model parameters in isolation. We need adaptable, high-agency problem solvers who take total ownership. One day you might be evaluating embedding models or optimizing a vector retrieval architecture; the next, you could be configuring AWS queues, containerizing services with Docker, or jumping into React to connect a backend service to the user interface. If you possess relentless grit, a "whatever it takes to ship" attitude, and want to learn directly from elite researchers and engineers, you will fit right in. What You’ll Be Doing * Next-Gen AI \& Verification Systems: Help design, build, and evaluate production-grade LLM workflows, Vector Retrieval systems, and custom embedding pipelines to power deterministic text verification. * Complex Data Parsing: Build pipelines to extract, parse, and structure messy, real-world document formats (PDF, XML) to preserve source integrity. * Backend Architecture: Write clean, scalable Python microservices, interact with databases, and handle asynchronous data flows via cloud message queues. * Vector Search \& Infrastructure: Work with high-dimensional vector search engines (e.g., Milvus), containerize applications using Docker, and support cloud infrastructure on AWS. * Full-Stack Integration: Implement telemetry for model performance and system health, while collaborating on frontend components in React. What We’re Looking For 1. The Right Attitude \& Location (Non-Negotiable) * Laser Focus \& Commitment: A high-agency mindset with relentless grit and a deep dedication to solving fundamental AI problems in an early-stage startup environment. * Cambridge-Based: You must be currently based in the Cambridge area and able to work on-site alongside the team to maintain high-velocity momentum. *(Only Cambridge-based applicants will be considered).* * Adaptability: Comfort moving fluidly between AI implementation, backend engineering, cloud tooling, and lightweight UI adjustments. 2. Technical Foundation * AI \& NLP: Foundational understanding of LLMs, RAG architectures, embeddings, and vector databases. * Software Engineering: Strong Python fundamentals, basic SQL knowledge, and familiarity with data formats (JSON, XML, PDF parsing). * DevOps \& Cloud: Exposure to Docker, microservices, and basic cloud platforms (AWS). * Frontend: Basic familiarity with React (or a strong willingness to pick it up quickly). Why Join FactTrace? * Learn from the Best: Work directly alongside top talent from the Cavendish Laboratory at the University of Cambridge. * Ground-Floor Impact: Join an early-stage startup tackling the defining challenge of the AI era—ensuring trust, traceability, and truth in automated information. * Rapid Growth: Accelerate your career at a pace impossible in traditional corporate environments through direct ownership of core systems.