Senior Machine Learning Engineer

Wiser Solutions is hiring a Senior ML Engineer to build its customer-facing AI platform and launch features for product matching, classification, and attribute extraction. The role owns ML systems end to end, from prototyping and evaluation through production deployment and ongoing improvement.

Responsibilities:

• Take customer problems from idea to working prototype quickly, prove or kill an approach with data, and move what works into production
• Train, fine-tune, and evaluate models for matching, classification, and extraction problems. This covers classical ML, embedding models, and LLMs, and you pick the simplest approach that meets the quality and cost bar
• Design and ship RAG systems, embedding-based retrieval, and LLM-powered features on AWS
• Own the full ML lifecycle: experiment tracking, model versioning, deployment, monitoring, and retraining
• Build evaluation that tells us whether something works: labeled golden sets, offline metrics (precision, recall), LLM-as-judge where it fits, and shadow or A/B testing before rollout
• Monitor models in production for drift, output quality, latency, and cost per prediction, and fix what you find

Requirements:

• 7+ years in ML engineering, applied data science, or AI engineering
• A track record of shipping ML or AI systems to production and owning them after launch. You can walk us through the architecture, the tradeoffs, and the metrics that showed it worked
• Hands-on experience training and evaluating models: feature engineering, experiment design, choosing metrics, handling class imbalance and noisy labels, and knowing when a model is good enough to ship
• Hands-on experience building production applications with LLMs, including RAG, embeddings, and vector databases
• Working knowledge of MLOps practices and tooling, such as MLflow, SageMaker Pipelines, Weights & Biases, or similar
• Experience with AWS for ML workloads (SageMaker, Bedrock, Lambda, ECS/EKS, S3)
• Strong Python and solid engineering habits: version control, testing, CI/CD, and code review
• A habit of building quickly to learn: you'd rather test an idea in two days than debate it for two weeks
• Daily use of AI coding assistants (Cursor, Claude Code, GitHub Copilot, or similar) and curiosity about new AI tools as they come out
• Experience with LLM fine-tuning, and a clear sense of when to fine-tune versus prompt engineer versus train a smaller task-specific model
• Experience keeping inference cost down at high volume through distillation, quantization, caching, or tiered model routing
• Background in retail, e-commerce, or product data (catalog matching, taxonomy mapping, attribute extraction)
• Experience with entity resolution or similarity search at scale
• Comfort working with data pipelines and orchestration tools like Airflow (helpful, not required)