Job Description & Details
This role sits at the intersection of traditional software engineering and cutting-edge generative AI, focusing on shipping actual production-grade LLM features rather than just running Jupyter notebooks. If you enjoy building scalable backends that interface with complex agentic workflows and RAG pipelines, this position gives you the ownership to do so.
What You'll Actually Be Doing
You'll spend your days designing and deploying AI-powered capabilities into enterprise applications, which means dealing with real-world latency, token costs, and model hallucinations. You will build out robust Retrieval-Augmented Generation (RAG) pipelines, set up automated testing and evaluation frameworks to measure model quality, and occasionally touch the React frontend to wire up your AI features. The core challenge here is moving beyond proof-of-concept models into reliable, production-ready distributed systems.
The Core Tech Stack
Python is your primary weapon here, and you'll need deep familiarity with building production-grade APIs and microservices. Experience with LLMs, prompt engineering, agentic workflows, and vector databases is non-negotiable since you'll be building enterprise-grade RAG applications. On top of that, you need a solid grasp of MLOps practices, CI/CD pipelines, and cloud platforms like AWS, Azure, or GCP to ensure your workloads scale cleanly and stay cost-efficient.
Interview Expectations
Expect to be grilled on how you handle failure modes in RAG systems, specifically regarding context window limits, grounding, and mitigating hallucinations in production. The hiring manager is going to want to walk through a distributed system you've built in Python, so be ready to explain how you handled API latency, error handling, and model monitoring under heavy enterprise load.
Application Advice
Your resume needs to clearly separate your traditional software engineering chops from your AI/ML specific projects. Don't just list buzzwords like "LLMs" and "Python"; highlight quantifiable impacts such as how much you reduced API latency, optimized token costs, or scaled an MLOps pipeline. Make sure terms like RAG, agentic workflows, evaluation frameworks, and CI/CD are explicitly mentioned to get past the initial ATS filters.