Job Description & Details
This is a solid production-focused AI role where you will actually be shipping machine learning and LLM features rather than just running endless Jupyter notebooks. If you enjoy building scalable pipelines for video entertainment distribution and want to work heavily with RAG and modern LLM APIs, this gig is worth your radar.
What You'll Actually Be Doing
You will spend your days designing and optimizing retrieval-augmented generation pipelines, wrangling embedding stores, and ensuring LLM outputs are reliably structured for downstream consumption. You'll be bridging the gap between raw experimental AI models and robust production systems, meaning you'll deal with prompt engineering, tool calling, and latency troubleshooting when APIs slow down. Expect to collaborate closely with product and data teams to explain model trade-offs without drowning everyone in academic jargon.
The Core Tech Stack
You need to be completely comfortable working with LLM APIs, prompt engineering, and building RAG systems utilizing modern embedding and search stores. Beyond the AI layer, strong foundational software engineering is non-negotiable—you'll need solid SQL skills, containerization with Docker, CI/CD pipelines, and cloud services. Clean modular design, automated testing, and solid observability practices will separate you from candidates who only know how to script prototypes.
Interview Expectations
Expect the hiring team to grill you on failure modes in RAG systems, such as how you handle context window limits and hallucinations when retrieving irrelevant chunks from your vector database. They are looking to see if you've actually debugged production AI apps or if you've just built toy demos. You will also likely be asked to break down a system design problem involving low-latency LLM responses and explain how you would monitor model performance and API costs in production.
Application Advice
Make sure your resume doesn't just list every AI buzzword under the sun; focus heavily on the 'to production' part of your background. Explicitly highlight your hands-on experience with vector search, RAG architectures, and API integrations. Use keywords directly from the description like 'prompt and context design', 'tool/function calling', and 'observability practices' to ensure you clear the ATS, and back those keywords up with metrics showing how you scaled or optimized a real-world application.