Data Scientist

This is a fully remote position, open to applicants in Peru.

📋 Description

• Develop and assess machine learning methodologies for matching companies/entities.

• Create matching techniques based on embeddings and large language models (LLM).

• Formulate scoring and ranking systems to identify genuine matches while differentiating them from duplicates, lookalikes, and unrelated entities.

• Handle complex datasets, which include names, aliases, domains, websites, firmographic details, multilingual records, and data hierarchies.

• Establish benchmark datasets, metrics, baselines, and processes for error analysis.

• Design and carry out experiments to test hypotheses.

• Evaluate LLM-assisted methods against more cost-effective alternatives.

• Analyze model performance, edge cases, and trade-offs.

• Consider the economics of inference and scalability from the outset.

• Share experimental results and recommendations with engineering and business stakeholders.

• Independently create experimental workflows and research strategies.

• Thoroughly document both successful and unsuccessful experiments.

⛳️ Requirements

• Over 5 years of professional experience in Data Science / Machine Learning.

• Solid foundation in applied Machine Learning principles.

• Proficient in Python and SQL.

• Practical experience with embeddings and semantic similarity.

• Hands-on application of LLMs to real-world challenges.

• Experience with both supervised and unsupervised learning techniques.

• Strong background in classification and natural language processing (NLP).

• Familiarity with neural networks and transformer architectures.

• Proficient in using TensorFlow, PyTorch, PyCaret, or similar ML frameworks.

• Experience in retraining or managing classification models in a production environment.

• Strong skills in experimental design and model evaluation.

• Experience in defining baselines, metrics, test sets, and conducting error analysis.

• Ability to assess model quality and demonstrate quantifiable enhancements.

• Deep understanding of scalability and costs associated with ML inference.

• Excellent communication skills in English.

• Nice-to-have: experience in entity resolution, record linkage, or deduplication.

• Nice-to-have: familiarity with ranking and similarity scoring.

• Nice-to-have: knowledge of retrieval, clustering, or candidate-generation techniques.

• Nice-to-have: LLM/embedding solutions tailored for cost and scalability constraints.

• Nice-to-have: experience with Spark, Snowflake, Databricks, or BigQuery.

• Nice-to-have: background in company, domain, website, or firmographic data.

• Nice-to-have: experience in working with multilingual datasets.

• Strong analytical and experimental mindset.

• Intellectual honesty and a willingness to communicate unfavorable results.

• High level of autonomy and self-direction.

• Excellent written and verbal communication abilities.

• Capability to defend technical recommendations to stakeholders.

• Strong problem-solving skills.

• Comfort with ambiguity and large-scale datasets.

• Ability to balance model quality, cost, and scalability.

🏝️ Benefits

• Option for remote work.