AI Engineer – Automated Data Transformation & Schema Mapping
Company Description Triple I is a leading provider of AI-powered tools for ESG reporting automation. The company offers solutions that handle the entire ESG process, from real-time data integration to audit-ready reports aligned with industry regulations. Trusted by teams across various industries, Triple I simplifies ESG reporting to help enterprises move faster, stay compliant, and reduce workloads. Role Description We’re looking for a skilled AI Engineer to build a powerful AI-driven system that can analyze, transform, and standardize raw datasets into a predefined destination schema — with full language normalization, schema mapping, and intelligent data validation. This role is perfect for someone with deep expertise in data pipelines, NLP, and intelligent schema inference who thrives on creating scalable, adaptable solutions that go far beyond hardcoded logic. What You’ll Be Doing Develop a generalizable AI algorithm that transforms raw, unstructured (or semi-structured) source datasets into a standardized schema Automate schema mapping, data enrichment, PK/FK handling, language translation, and duplicate detection Build logic to flag unresolved data, generate an UnresolvedData_Report, and explain confidence or failure reasons Ensure all outputs are generated in English only, regardless of input language Experiment with 2–3 AI/ML approaches (e.g. NLP models, rule-based logic, transformers, clustering) and document tradeoffs Deliver all outputs (destination tables) in clean, validated formats (CSV/XLSX) Maintain detailed documentation of preprocessing, validation, and accuracy logic Key Responsibilities Design AI logic to dynamically extract, map, and organize data into 10+ destination tables Handle primary key/foreign key relationships across interconnected tables Apply GHG Protocol logic to assign Scope 1, 2, or 3 emissions automatically based on activity type Build multilingual support: auto-translate non-English input and ensure destination is 100% English Handle duplicate and conflicting records with intelligent merging or flagging logic Generate automated validation logs for transparency and edge case handling
