Job Description:
We are currently seeking a Data Migration AI Engineer to join our team in Bangalore, Karnātaka (IN-KA), India (IN). "Job Duties: Role Overview The Data Migration AI Engineer bridges NTT DATA's Data as Agentic Product (DaaP) platform and the hands-on execution of SEI's Informatica-to-dbt migration program. This role operates in two sequential phases, each with a distinct mandate. 1. In Phase 1, the AI Engineer works closely with the Onshore Technical Lead and dbt SMEs to perform context engineering — crafting, iterating, and validating the prompts, agent configurations, and system context that guide DaaP's code generation agents toward producing Python EL pipelines and dbt transformation models that conform to SEI's architecture standards, naming conventions, and data modeling patterns. The goal of Phase 1 is a validated, repeatable DaaP configuration that consistently produces migration-ready code output with minimal rework. 2. Once the DaaP configuration is deemed satisfactory, the AI Engineer transitions into Phase 2 — active participation in the migration execution itself. In this phase, the AI Engineer applies AI-assisted generation to accelerate the conversion of Informatica mappings, sessions, and workflows into Python extract/load scripts and dbt SQL models, working alongside Data Engineers, Informatica SMEs, and dbt SMEs across the SWP and IMS migration workstreams. Key Responsibilities Prompt & Agent Configuration • Design, iterate, and validate prompt templates that guide DaaP agents to produce Python EL and dbt code aligned with SEI's architecture and standards. • Configure DaaP's Data Discovery, Data Mapping, and ETL & Code Review agents for the SEI migration context. • Establish system context, few-shot examples, and output constraints that enforce SEI's coding conventions, model layering (staging → intermediate → marts), and dbt testing patterns. • Define and document the context engineering artifacts (prompts, agent configs, example inputs/outputs) used to tune DaaP output. Output Validation & Iteration • Evaluate DaaP-generated Python EL scripts and dbt models against agreed quality criteria in collaboration with the Onshore Technical Lead and dbt SMEs. • Identify failure modes, hallucinations, and structural deviations in generated code and iterate on context configuration to resolve them. • Establish a validation gate — a defined set of criteria the DaaP output must meet before Phase 2 execution begins. • Document context engineering decisions, prompt versions, and validation results for program governance and knowledge transfer. • Monitor DaaP output quality across migration waves and refine context configurations as new Informatica complexity patterns emerge. AI-Assisted Code Generation • Apply DaaP and validated prompt configurations to accelerate conversion of Informatica mappings and sessions into Python EL scripts. • Generate dbt model scaffolding (staging, intermediate, marts) from Informatica transformation logic in collaboration with Informatica SMEs. • Review and refine AI-generated code before handoff to Data Engineers for integration testing and validation. • Identify patterns in Informatica constructs (lookups, aggregators, routers, update strategies) that benefit from AI-assisted translation and develop reusable generation templates. • Work with Informatica SMEs to accurately capture source transformation logic as structured input context for DaaP agents. • Partner with dbt SMEs to ensure generated dbt models conform to architectural standards and pass code review. Continuous Improvement • Monitor DaaP output quality across migration waves and refine context configurations as new Informatica complexity patterns emerge. • Maintain a prompt and configuration library versioned in GitLab alongside migration artifacts. • Contribute to post-migration documentation and AI tooling runbooks for steady-state use.
