AI/ML Engineer

Role Summary

We are seeking a Mid-level AI/ML Engineer to support the design, training, and optimization of the machine learning models that power Nutpaa’s current and future AI products. You will focus on the core execution work of ML: preparing data, implementing training pipelines, running experiments, tracking results, and helping move models towards deployment readiness.

This role is ideal for someone with 3–4 years of hands-on ML experience who wants to deepen their skills.

Key Responsibilities

1. Data Preparation & Dataset Management

• Help design and maintain datasets for different AI products (vision, sequence, or multimodal), including:

• Data collection, cleaning, and basic exploratory analysis.

• Applying and maintaining labeling schemas defined by senior engineers.

• Implement and maintain data pipelines for:

• Train/validation/test splits

• Versioning datasets and keeping track of changes.

• Ensuring basic data quality checks (missing values, outliers, label consistency).

2. Data Augmentation & Preprocessing

• Implement augmentation strategies defined in collaboration with senior engineers (e.g., geometric and photometric transforms for images, noise/perturbations for other data types).

• Build preprocessing scripts and utilities for:

• Normalization, scaling, filtering.

• Sequence preparation (windowing, padding, temporal smoothing) where applicable.

• Analyze the impact of preprocessing and augmentation on validation metrics and share findings.

3. Model Training & Experiment Support

• Implement and maintain training loops and model definitions (primarily in PyTorch).

• Run experiments under the guidance of senior engineers, including:

• Hyperparameter sweeps (learning rate, batch size, regularization).

• Architecture variations within agreed ranges.

• Monitor training and validation metrics (loss curves, accuracy curves, etc.) and flag anomalies early.

• Debug common training issues (e.g., NaNs, diverging loss, data/label mismatches).

4. Evaluation, Analysis & Reporting

• Implement evaluation scripts to compute agreed metrics (accuracy, precision/recall, F1, regression metrics, etc.).

• Perform basic error analysis (e.g., inspect misclassified samples, problematic edge cases).

• Summarize experiment results in concise reports or dashboards for the team.

• Contribute to defining simple but meaningful benchmarks that reflect real-world performance.

5. Reproducibility, Experiment Tracking & Documentation

• Use experiment tracking tools (e.g., metric dashboards, run logs) to record:

• Configurations, hyperparameters, dataset versions, and code versions.

• Ensure key experiments are reproducible by:

• Fixing random seeds.

• Documenting dependencies and environment details.

• Maintain clear documentation for:

• Training scripts and utilities.

• Data preprocessing steps.

• Experiment results and conclusions.

6. Deployment Readiness Support

• Collaborate with senior engineers and the Mobile ML Engineer to:

• Export trained models to deployment formats (e.g., ONNX, TFLite, CoreML, etc.) as needed.

• Prepare evaluation datasets for quantization/optimization checks.

• Participate in discussions about latency, memory, and accuracy trade-offs and help run tests to measure these impacts.

7. Collaboration & Continuous Learning

• Work closely with senior ML and CV engineers, taking feedback on code, experiments, and approach.

• Participate in technical discussions, design reviews, and retrospectives.

• Proactively learn new techniques relevant to current projects (reading papers, trying small prototypes).

8. RAG & Agent Pipeline

• RAG Pipeline Development – Build and maintain retrieval pipelines and embedding models that power domain-specific RAG for coaching and analytics.

• Agent Training Pipeline – Design simple agents to monitor data quality, detect labeling issues, and suggest improvements to datasets and training jobs.

Required Skills & Experience (Junior Level)

Educational Background

• Bachelor’s degree in a relevant field such as: Computer Science, Mathematicsm Statistics, Physics, Electrical/Electronics/Computer Engineering

• Other degrees considered if you can demonstrate strong foundations in math and programming.

Core Technical Skills

• 3–4 years hands-on experience with machine learning (industry, research lab, or substantial projects).

• Strong Python skills, including:

• NumPy, SciPy, pandas for numerical computing and data manipulation.

• Proficiency in data visualization (Matplotlib, Seaborn, or similar).

• Practical experience with at least one deep learning framework, preferably: PyTorch (primary), and/or TensorFlow/Keras.

• Experience building and training models on standard tasks (classification, regression, or simple CV tasks).

Machine Learning Fundamentals

• Solid understanding of:

• Train/val/test splits, overfitting vs. underfitting.

• Common loss functions (MSE, cross-entropy) and when to use them.

• Gradient descent and basic optimizers (SGD, Adam).

• Regularization concepts (dropout, weight decay, early stopping).

• Experience training small-to-medium models end-to-end and improving them through iteration.

Data & Experimentation Skills

• Experience working with real datasets (not just toy examples), including cleaning, transforming, and augmenting data.

• Comfort with:

• Writing data loaders and preprocessing pipelines.

• Logging metrics and saving model checkpoints.

• Familiarity with at least one experiment tracking approach (from simple spreadsheets/notebooks to dedicated tools).

Mathematical Foundation

• Good working knowledge of:

• Linear algebra: vectors, matrices, basic decompositions.

• Calculus: derivatives, chain rule, gradient concepts.

• Probability & statistics: distributions, expectation/variance, basic hypothesis testing.

• Ability to connect mathematical concepts to model behavior and training dynamics.

Communication & Code Quality

• Ability to write clear, modular, well-documented code.

• Comfortable explaining your experiments, results, and reasoning to other engineers.

• Willingness to iterate based on feedback and to improve code and documentation quality.

Agentic Frameworks and RAG systems

• RAG Systems – Vector DBs, embedding selection/fine-tuning, chunking strategies, retrieval quality evaluation.

• Agent Frameworks (Basic) – Using agent frameworks to automate data checks, labeling assistance, and experiment monitoring.

Preferred Skills & Experience (Not required, but strong pluses.)

• Experience with computer vision tasks (image classification, detection, pose, etc.).

• Participation in Kaggle or similar competitions.

• Experience with experiment tracking tools (e.g., Weights & Biases, MLflow, TensorBoard).

• Exposure to model deployment or format conversion (ONNX, TFLite, etc.).

• Contributions to ML-related open-source projects.

• Experience reading and implementing ideas from ML/AI research papers.

What You’ll Gain

• Deep exposure to end-to-end ML workflows in a real product environment.

• Close mentorship from senior ML and CV engineers on both theory and practice.

• Opportunity to ship models that power real products used across different domains.

• A structured path to grow into a mid-level and then senior AI/ML Engineer role.

• Hybrid working opportunities post-MVP phase (Month 6 onwards), with flexibility for remote collaboration aligned with team needs.

Organizational & Cultural Expectations

• Bring curiosity, humility, and a willingness to learn quickly.

• Maintain rigor in experiments and be honest about what works and what doesn’t.

• Engage constructively in code reviews and technical discussions.

• Uphold Nutpaa’s values of Engineering Excellence, Long-Termism, Open Evolution, and Peer-Driven Collaboration.

Application Process

Please email careers@nutpaa.ai with:

1. Resume highlighting ML/AI experience, projects, and education.

2. Portfolio (if available), such as:

• GitHub repositories with ML projects.

• Links to any blog posts, notebooks, or write-ups explaining your work.

• Kaggle profile or similar competitive work.

3. A short statement of interest (~150–200 words) explaining:

• Why you want to work on production AI systems at Nutpaa.

• One ML project you’re proud of and what you learned from it.

Email Subject: Junior AI/ML Engineer – Model Training & Optimization – [Your Name]

Equal Opportunity Statement

Nutpaa is an equal opportunity provider. We do not discriminate based on race, religion, color, national origin, gender, gender identity or expression, sexual orientation, age, marital status, veteran status, or disability status.