Senior AI​/Machine Learning Engineer

The Elevator Pitch

Join Evolv as Senior AI/Machine Learning Engineer to advance AI innovation in physical security technology. As a key team member of the AI/ML team, you will be developing and deploying state‑of‑the‑art machine learning and deep learning solutions. Your role will involve leveraging diverse data sources, including magnetic sensors, 3D cameras and other sensors, to create multi‑sensor fusion solutions that operate in real‑time on constrained hardware platforms.

This hands‑on role requires deep expertise in classical ML, deep learning, feature engineering, model optimization, and MLOps. You will drive modeling strategy, strengthen model accuracy and robustness, and deploy reliable models in real‑world environments. This position is ideal for someone known for measurably improving models—not just building them.

Success in the Role:
What performance outcomes will you work toward in the first 6–12 months?

In the first 30 days:

• Learn the sensor ecosystem, ML pipelines, and development standards.
• Review real‑time constraints, production workflows, and existing model performance baselines.
• Engage in code reviews and collaborate across engineering teams.
• Identify key opportunities for improving accuracy, latency, and robustness.

Within the first three months:

• Lead feature engineering from raw sensor inputs, including temporal, spectral, and statistical features.
• Develop and optimize classical ML and deep learning models.
• Propose model improvements through systematic experimentation and benchmarking.
• Partner with product and hardware teams to translate sensor behavior into ML architectures.

By the end of the first year:

• Own end‑to‑end ML model lifecycle for core production systems.
• Deploy scalable ML models and ensure operational reliability.
• Drive architecture decisions balancing classical ML and deep learning approaches.
• Improve robustness across devices and field environments by modeling sensor characteristics.

The Work:
What type of work will you be doing? What assignments, requirements, or skills will you be performing on a regular basis?

Technical Leadership:

• Design, develop, and optimize ML models—including XGBoost, Random Forests, SVMs, CNNs, and Transformers.
• Lead hyperparameter tuning, feature selection, and algorithm evaluation.
• Integrate models into production system, work with SW team on optimizing runtime speed and performance.
• Develop reproducible training pipelines with model, data, and experiment versioning.
• Feature Engineering & Sensor‑Aware Modeling.
• Extract temporal, spectral, and domain‑specific features from raw sensor signals.

• Use data analytics tools such as UMAP and T‑SNE to understand data distribution and feature characteristics.
• Model sensor characteristics such as noise, bias, drift, and environmental effects.
• Perform ablation studies and feature importance analyses (SHAP, PDP, etc.).
• Multi‑Class Detection & Classification:
• Design multi‑class object detection and classification pipelines for noisy, imbalanced datasets.
• Define evaluation metrics including confusion matrices, calibration, and class‑wise scoring.
• MLOps & Production Excellence:
• Deploy production‑ready ML code impacting real customers.
• Ensure reliability through CI/CD, drift detection, and data validation.
• Optimize models for edge and compute‑constrained environments.
• Cross‑Functional
  
  Collaboration:
• Work with hardware, software, product and cross‑functional teams.
• Communicate technical decisions and trade‑offs to senior stakeholders.

Minimum Qualifications:

• Master’s or PhD in Computer Science, Machine Learning, Engineering, Applied Math, Physics, or related field.
• 3‑5+ years building and deploying ML models for real‑world applications.
• Strong expertise in classical ML techniques (e.g., XGBoost, Random Forests, SVM, k‑NN) and modern ML techniques (e.g., deep neural network, transformers).
• Proficiency in Python, ML libraries (scikit‑learn, Num Py, pandas) and C++.
• Experience with multi‑class classification on real‑world, noisy datasets.
• Strong statistical and model evaluation skills.

Preferred Qualifications:

• Experience with sensor or time‑series data (magnetic, radar, 3D, IoT).
• Advanced feature extraction (FFT, windowing,…