(Remote) Data Scientist

📋 Description

• • Own the end-to-end lifecycle of machine-learning models that directly influence how millions of products are discovered in AI search engines. You will design, train, evaluate, and deploy models that translate raw search-behavior data into precise, revenue-driving recommendations for our customers.
• • Reverse-engineer the decision logic of leading LLMs (OpenAI, Claude, Gemini, Llama, Perplexity, etc.) to uncover the ranking signals that determine visibility. Translate these insights into new product features that give brands a measurable edge in AI-driven search.
• • Build and maintain scalable data pipelines on Google Cloud Platform (Cloud Functions, Firestore, BigQuery, AlloyDB, Postgres) that ingest terabytes of search-interaction logs, clean and enrich the data, and surface real-time features to downstream models.
• • Partner with backend engineers to productionize models via FastAPI micro-services, containerize with Docker, and orchestrate serverless deployments that auto-scale to handle traffic spikes without manual intervention.
• • Run rigorous A/B and multivariate experiments to quantify uplift in AI search visibility; present statistically significant findings to both technical and non-technical stakeholders, influencing the product roadmap with data-backed narratives.
• • Continuously monitor model drift, latency, and cost in production; implement automated retraining, canary releases, and rollback strategies that keep our systems reliable while shipping improvements weekly.
• • Contribute to the open-source community by releasing reusable components of our ML stack, writing technical blog posts, and speaking at meetups or conferences to establish Peec as a thought leader in AI search optimization.
• • Mentor junior data scientists and engineers, fostering a culture of reproducible research, peer code reviews, and shared ownership of our experimentation platform.
• • Collaborate directly with customers during beta programs, translating their raw business questions into well-scoped data problems and delivering dashboards or notebooks that let them see the impact of your models in real time.
• • Stay ahead of the rapidly evolving LLM landscape; prototype new techniques such as retrieval-augmented generation, fine-tuning, or prompt optimization, and rapidly validate which approaches move the needle for our users.
• • Champion ethical AI practices—audit models for bias, ensure GDPR-compliant data handling, and build transparency tools that let customers understand why a recommendation was made.
• • Shape the future architecture of our ML platform, evaluating cutting-edge tools (ONNX for model portability, TensorFlow Serving, PyTorch Lightning) and making build-vs-buy decisions that balance velocity, cost, and long-term maintainability.