Senior AI Engineer

We are looking for a hands-on AI Engineer to build, test, and improve AI-driven solutions for the publishing business. This is a practical technical role. We are not looking for someone who only talks about AI strategy, builds simple no-code automations, or experiments with visual app builders. We are looking for someone who can turn ideas into working software, test things quickly, and help move useful AI solutions toward production. The ideal candidate is technically strong, curious, independent, and comfortable working with emerging AI technologies. You should be familiar with LLMs, APIs, agentic workflows, and frameworks such as LangChain, LangGraph, LlamaIndex, or similar tools. You should also be comfortable using modern agentic coding tools such as Codex, Cursor, Claude Code, GitHub Copilot, or similar systems as part of your development workflow. What You Will Do Build, run, and evaluate proof-of-concept and MVP initiatives, especially within automated content creation and AI-assisted publishing workflows. Develop practical AI solutions using LLMs, APIs, orchestration frameworks, retrieval systems, automation pipelines, and agentic workflows. Experiment quickly with new tools, frameworks, models, and technical approaches. Translate business needs into working technical solutions with a focus on reliability, speed, and measurable value. Take ownership of assigned technical tasks and move work forward without constant supervision. Identify opportunities, blockers, and next steps proactively. Report progress clearly to the Tech Lead / AI Tech Manager and stay aligned with agreed priorities, processes, and reporting lines. Work closely with QA, product, editorial, and production teams to validate model outputs and improve reliability. Support the handover of validated PoCs and MVPs into production-ready solutions. Stay current with emerging AI frameworks, LLM capabilities, coding agents, evaluation methods, and AI engineering best practices. Strong hands-on experience with AI/ML tools, LLMs, API integrations, and AI application development. Experience with frameworks such as LangChain, LangGraph, LlamaIndex, or similar tools. Solid programming skills, preferably in Python and/or TypeScript. Practical understanding of prompt engineering, retrieval-augmented generation, agents, workflow automation, model evaluation, and output validation. Experience using agentic coding tools such as Codex, Cursor, Claude Code, GitHub Copilot, or similar tools to build real software. Ability to move quickly from idea to prototype while maintaining sound engineering judgment. Strong analytical skills and the ability to test model outputs, interpret feedback, and improve system design. Comfort with ambiguity, fast iteration, and unfamiliar technical problems. Ability to work independently while keeping managers informed and following agreed direction. Clear communication skills for cross-functional collaboration. Important Fit Note This is not a no-code or low-code automation role. Experience with tools such as Make, Zapier, Lovable, Bolt, or similar platforms can be useful, but it is not enough for this position on its own. We are looking for someone who can code, debug, integrate APIs, design AI-powered workflows, evaluate outputs, and build systems that can become reliable production tools. You should be comfortable working directly with code and using AI as part of a serious engineering workflow. Personal Profile Highly technical and hands-on. Bright, curious, and eager to try new things. Pragmatic and delivery-focused. Comfortable experimenting and learning independently. Takes initiative and can work without constant hand-holding. Accountable, communicative, and aligned with team direction. Motivated by building useful solutions, not excessive discussion or theory. Final Project Requirement Candidates must document something they have personally built where the final project or pipeline uses AI. As part of the final application process, candidates must present a small project, prototype, workflow, pipeline, or technical solution they have built. The final result must actively use AI, such as an LLM, AI API, agentic workflow, retrieval system, automation pipeline, model-based classification, content generation flow, or similar AI-driven component. The documentation should clearly show: What problem they were trying to solve. How AI is used in the final project or pipeline. Which AI tools, frameworks, APIs, models, or coding agents they used. How they approached the build. What worked, what failed, and what they changed along the way. The final result and how it could create practical value. Projects built only with no-code or low-code tools will not meet this requirement unless there is also clear technical implementation, coding, API integration, or system design involved. This requirement is intended to demonstrate hands-on ability, technical curiosity, independent execution, and comfort with using AI in real project work.