AI Engineer — Text-to-SQL, Data Agents & Voice AI

Riyalabs.ai

India

About RiyaLabsRiyaLabs helps organizations move enterprise AI from experimentation into dependable, governed business execution. We build AI coworkers, coordinated AI teams, enterprise integrations, and practical controls for responsible AI use in real workflows.

Our offerings include RiyaLabs Studio for AI coworkers and coordinated AI teams; RiyaLabs Trust Gateway for governance, approval, observability, and policy controls; RiyaLabs Consultants for AI strategy and implementation; and RiyaLabs Academy for practical AI capability building.

Role PurposeRiyaLabs is seeking an AI Engineer with strong experience in Text-to-SQL, relational databases, semantic layers, AI-powered analytics, and dynamic dashboards.

You will build data agents that allow business users to ask questions in natural language, understand trusted business definitions, safely query approved relational databases, and receive answers through dynamic dashboards, charts, tables, and narrative summaries.

This is not a prompt-engineering-only role. You will build reliable systems that understand database schemas, table relationships, business metrics, dimensions, access controls, query performance, and approved data boundaries.

Experience with Snowflake Cortex Agents, Cortex Analyst, Snowflake semantic models, or Snowflake-native AI capabilities is strongly preferred. Experience developing voice agents or voice-enabled AI applications is preferred.

Key Responsibilities:Text-to-SQL and Semantic Data Agents

Build AI data agents that convert natural-language business questions into accurate, secure, explainable SQL.
Develop semantic layers that map business terms, KPIs, metrics, dimensions, calculations, tables, joins, relationships, and access rules to underlying relational data.
Build schema-ingestion, metadata-enrichment, business-glossary, retrieval, and context-management capabilities for AI agents.
Implement query planning, SQL generation, validation, repair, approved execution, result explanation, and follow-up-question workflows.
Use semantic definitions, metadata, approved query patterns, business rules, user permissions, and validation controls—not only raw schema information.
Create evaluation datasets and regression tests for SQL validity, execution success, semantic correctness, answer accuracy, latency, safety, and cost.
Identify ambiguous requests, unclear metrics, inconsistent schemas, missing joins, data-quality issues, and other risks that may produce inaccurate results.

Dynamic Dashboards and Analytics

Generate dashboards, KPI cards, charts, tables, drill-downs, comparisons, alerts, and narrative summaries dynamically based on the user’s natural-language question and query results.
Build conversational analytics experiences that allow users to refine time periods, filters, segments, calculations, comparisons, and visualizations through follow-up questions.
Create transparent data experiences that show relevant SQL, data sources, filters, metric definitions, refresh timing, and applicable limitations.
Develop reusable analytics patterns for trends, period comparisons, segmentation, anomaly investigation, top/bottom analysis, and operational reporting.
Build data applications and visualizations using approved dashboard and embedded-analytics technologies.

Data Platform and Engineering

Write, optimize, review, and troubleshoot complex SQL involving joins, CTEs, window functions, aggregations, subqueries, views, and performance optimization.
Build secure backend services, APIs, connectors, and integration layers using Python and approved frameworks.
Work with relational databases including Snowflake, PostgreSQL, MySQL, SQL Server, Oracle, BigQuery, Redshift, Databricks SQL, or equivalent platforms.
Build solutions using Snowflake SQL, Snowpark, secure views, RBAC, warehouses, Streams, Tasks, Dynamic Tables, and data-sharing concepts where applicable.
Develop or support Snowflake Cortex Analyst, Cortex Agents, Cortex Search, semantic models, or comparable enterprise data-agent capabilities.
Integrate AI workflows with approved databases, APIs, documents, and enterprise systems using REST APIs, webhooks, OAuth, service accounts, and scoped permissions.
Optimize data-access patterns, query performance, caching, warehouse usage, and cost.

Voice AI, Security, and Delivery

Build or support voice-enabled AI experiences using speech-to-text, text-to-speech, conversational orchestration, session context, confirmations, and error recovery.
Route voice requests to approved AI data agents and require confirmation for sensitive or consequential actions.
Apply read-only data access, role-based permissions, tenant isolation, approved source boundaries, query limits, row limits, cost controls, timeouts, and audit logging.
Implement safeguards against prompt injection, SQL injection, unauthorized data access, data leakage, unsafe tool use, and insecure outputs.
Log and trace user requests, generated SQL, executed queries, results, agent decisions, model/tool calls, errors, latency, token usage, and cost.
Participate in code reviews, testing, security reviews, CI/CD, deployment, incident investigation, and reliability improvements.
Produce architecture diagrams, API specifications, semantic-model documentation, data-flow diagrams, test plans, deployment guides, and runbooks.

Required Skills

Strong SQL and relational-database expertise.
Experience building, debugging, reviewing, and optimizing complex SQL queries.
Strong Python and backend-development experience using FastAPI, Flask, Django, or equivalent frameworks.
Experience with Text-to-SQL, natural-language analytics, AI data agents, semantic layers, RAG, LLM applications, or conversational analytics.
Experience designing schemas, joins, metrics, dimensions, data models, business definitions, and secure data-access patterns.
Experience building dashboards, analytics applications, embedded reporting, data APIs, or BI solutions.
Familiarity with OpenAI, Anthropic, Gemini, Azure OpenAI, or approved open-source models.
Understanding of prompting, structured outputs, function calling, tool use, agent orchestration, evaluation, and output validation.
Familiarity with LangChain, LangGraph, LlamaIndex, Semantic Kernel, DSPy, or equivalent frameworks.
Experience with REST APIs, webhooks, OAuth 2.0, JWTs, RBAC, service accounts, Git, Docker, CI/CD, Linux, and cloud deployment.
Familiarity with AWS, Azure, GCP, or comparable cloud platforms.
Understanding of database and AI security risks, including SQL injection, prompt injection, data leakage, excessive permissions, and unsafe tool use.
Familiarity with OpenTelemetry, LangSmith, Grafana, Prometheus, Datadog, Sentry, Snowflake query history, or equivalent observability tools.

Required Experience

3–6 years of experience in software engineering, data engineering, analytics engineering, AI engineering, backend engineering, or related technical roles.
Production experience with relational databases and complex SQL.
Experience building dashboards, data applications, embedded analytics, data APIs, or BI solutions.
Experience building LLM-powered applications, Text-to-SQL systems, RAG solutions, AI data agents, automation workflows, or conversational AI.
Ability to translate business questions into semantic definitions, technical designs, and production-ready solutions.
Strong written and verbal English communication skills.

Preferred Qualifications

Snowflake Cortex Agents, Cortex Analyst, Cortex Search, Snowpark, Snowflake semantic models, or Snowflake-native AI experience.
Voice-agent, speech-to-text, text-to-speech, telephony, real-time streaming, or conversational voice-AI experience.
Experience with Streamlit, React, Next.js, TypeScript, Plotly, Power BI, Tableau, Looker, Metabase, Apache Superset, or embedded analytics.
Experience with dbt, data catalogs, semantic layers, metrics stores, data lineage, governance, or data-quality tools.
Experience with ERP, CRM, HRIS, finance, operations, supply-chain, or customer-support data.
Experience with multi-tenant SaaS, enterprise identity and access management, Kubernetes, Terraform, cloud DevOps, or API gateways.
Technical portfolio, GitHub projects, published dashboards, AI/data-agent prototypes, or open-source contributions.

Tools and Technology EnvironmentProgramming: Python, SQL, TypeScript/JavaScript

Databases: Snowflake, PostgreSQL, MySQL, SQL Server, Oracle, BigQuery, Redshift, Databricks SQL

AI and Text-to-SQL: OpenAI, Anthropic, Gemini, Azure OpenAI, LangChain, LangGraph, LlamaIndex, Semantic Kernel, DSPy, structured outputs, tool calling, and agent orchestration

Snowflake: Snowflake SQL, Snowpark, Cortex Analyst, Cortex Agents, Cortex Search, semantic models, secure views, RBAC, warehouses, Streams, Tasks, and Dynamic Tables

Dashboards: Streamlit, React, Next.js, Plotly, Power BI, Tableau, Looker, Metabase, Apache Superset, and embedded analytics tools

Voice AI: OpenAI Realtime API, Twilio, Deepgram, ElevenLabs, AssemblyAI, Google Speech-to-Text, Azure Speech, Amazon Transcribe/Polly, LiveKit, or equivalent

Cloud and Observability: AWS, Azure, GCP, Docker, Kubernetes, OpenTelemetry, LangSmith, Grafana, Prometheus, Datadog, and Sentry

Success Measures

Accurate, safe, explainable, and performant Text-to-SQL output.
Strong semantic mapping, metric definitions, schema understanding, and access controls.
Trusted, responsive, dynamically generated dashboards and conversational analytics.
Effective testing, evaluation, monitoring, documentation, and operational reliability.
High-quality code, reusable components, technical collaboration, and responsible handling of client data.

Score my resume against this job, free →