Job Description & Details
The role is a senior AI/ML Engineer gig in Indianapolis, geared toward someone with a decade of hands‑on experience. It’s a contract position, so you’ll be expected to hit the ground running and deliver production‑ready models without a lot of onboarding fluff.
What You'll Actually Be Doing
You’ll spend most of your day designing, training, and deploying machine‑learning pipelines that solve concrete business problems. Expect to clean noisy data, experiment with feature engineering, and iterate on model architectures while keeping latency and scalability in mind. The contract nature means you’ll also need to document your work clearly and hand it off to the client’s team at the end of the engagement.
The Core Tech Stack
The stack is all about Python‑centric ML: you’ll be deep in NumPy, pandas, and scikit‑learn for classic models, and TensorFlow or PyTorch for deep learning work. The client also runs models in a cloud environment (likely AWS or Azure), so containerizing with Docker and orchestrating with Kubernetes is a must. They care about reproducibility, so version control with Git and experiment tracking (MLflow, Weights & Biases) are non‑negotiable.
Interview Expectations
- “Walk me through how you’d handle concept drift in a production model that’s seeing a gradual shift in input data distribution.” They want to see you talk about monitoring pipelines, statistical tests, and a plan for periodic retraining or online learning. 2. “Explain the trade‑offs between using a large transformer model versus a distilled version for real‑time inference.” Expect them to probe your understanding of latency, GPU memory constraints, and how distillation can keep accuracy while cutting cost.
Application Advice
Tailor your resume to highlight the exact buzzwords the JD drops: “AI/ML Engineer,” “10+ years,” “Python,” “TensorFlow/PyTorch,” “Docker,” “Kubernetes,” and “AWS/Azure.” Put those terms in your experience bullet points and in a skills summary so the ATS flags you. Also, mention any contract work you’ve done where you delivered end‑to‑end pipelines, because that signals you can thrive in a short‑term, results‑focused setting.