The Mission: Build the Brain of an Autonomous System

At Zode, we believe the future of production reliability is autonomous. We are building the world's first autonomous governor for CI/CD—an intelligent layer that predicts incidents before they happen and takes action to prevent them.

Our core promise rests on one thing: a predictive model that is not just accurate, but honest, explainable, and incredibly fast.

We are looking for our Founding AI/ML Engineer. This isn't a role for a researcher. This is a role for a builder. You will be the architect and owner of the "conscious" part of the Zode the brain that turns chaotic, high-dimensional time-series data into a single, trusted decision. This is a chance to build a category-defining ML system from the ground up, solving problems that directly prevent outages and eliminate on-call pain for fellow engineers.

What we Will Build & Own:

• The Core Predictive Engine: You will design, build, and productionize the models that forecast SLO breaches, p95 latency spikes, and capacity exhaustion. This is your canvas, from feature engineering to model serving.

• A Modern ML Stack: You will own the end-to-end ML pipeline. Our current thinking is a stack built on Python, PyTorch/Scikit-learn, and serving with ONNX Runtime, but you will be the one to make the final calls and build it for scale.

• Sophisticated Modeling Techniques: This isn't just another classification problem. You'll be working with multi-horizon time-series forecasting (TCNs, LSTMs), robust feature engineering, and applying techniques like Conformal Prediction to provide statistically honest confidence intervals—no black boxes.

• Causal & Explanatory AI: A prediction is useless without a reason. You will build the systems that provide attribution, linking a predicted risk back to a specific change marker and a set of contributing factors.

• A System That Learns: You will design the feedback loops that allow our models to learn from manual overrides and real-world outcomes, creating a flywheel that makes the entire system smarter with every deployment it governs.

Who You Are:

• A Pragmatic Builder: You have a proven track record of shipping production ML systems. You are comfortable with the entire lifecycle, from wrangling data with SQL/Pandas to serving a model behind a low-latency API.
• A Time-Series Expert: You have deep, hands-on experience with time-series forecasting. You understand concepts like seasonality, stationarity, and the pros and cons of different modeling approaches (from classical methods to deep learning).
• Product-Minded: You are obsessed with the "so what?" of a prediction. You constantly think about the end-user, the actions they will take, and how to build a system that earns their trust.
• Technically Fluent: You are strong in Python and its ML ecosystem (PyTorch, TensorFlow, Scikit-learn, etc.). You have experience with MLOps principles and tools. Experience with Go is a plus but not required.
• An Independent Owner: As a founding engineer, you are comfortable with ambiguity. You can take a high-level architectural spec, debate its merits, and then go execute on it with a high degree of autonomy.

The Opportunity:

This is more than a job. It's a chance to be a true owner in a company that is solving a massive, universal problem for engineers. You will have a significant equity stake, a huge amount of technical ownership, and the opportunity to build a product and a team that will define the future of reliable software delivery.

If you are an ML engineer who wants to build something that actively prevents outages instead of just analyzing them, we want to talk to you.