Permanent

Position: Permanent
Role: Data Engineer
Experience: 3 – 6 years
Work Location: Hyderabad / Bangalore
Notice Period: Immediate to 30 days
Type: Full-time
Job Summary:
We’re looking for a Data Engineer to support the development of robust data evaluation pipelines and frameworks
for assessing the performance of natural language and SQL-based systems. You’ll play a key role in designing
scalable data tools, evaluating complex SQL outputs, and contributing to the continuous improvement of machine
learning model quality.
This role is ideal for engineers with strong SQL expertise, analytical rigor, and experience working at the
intersection of data infrastructure and ML system evaluation.
Key Responsibilities:
• Design and enhance data evaluation pipelines for testing and validating SQL-generating models
• Develop, verify, and optimize SQL queries across multiple dialects including PostgreSQL, MySQL,
GoogleSQL, and SQL Server
• Collaborate with research and engineering teams to manually evaluate model-generated SQL results and
identify correctness issues or edge cases
• Assist in the development of error classification systems and quality metrics for SQL outputs
• Build reusable scripts, tools, or dashboards to automate reporting and highlight accuracy, quality, and
regression trends
• Contribute to defining test datasets and supporting multi-turn or iterative query workflows
• Provide input into tool and framework design decisions to support scalable evaluation of language model
outputs
Qualifications:
Must:
• 3+ years of Strong proficiency in SQL, including writing, debugging, and optimizing complex queries
• Experience working with multiple SQL dialects (e.g., PostgreSQL, MySQL, GoogleSQL, SQL Server)
• Proficiency in Python, especially for scripting evaluations and building data tools
• Experience in manual validation of data or model outputs, and an eye for detail when inspecting edge
cases
Nice to Have:
• Experience with natural language to SQL (NL2SQL) or query generation tasks
• Knowledge of LLM-based applications, dialog systems, or prompt engineering
• Familiarity with concepts around machine learning model evaluation, including correctness,
robustness, and interpretability
• Familiarity with database schema design and synthetic or benchmark dataset generation