Software Engineer

Allstate India Private Limited; Pune District, Bengaluru; On-site; INR 1,800,000 - 2,600,000; Full time

Job summary

Allstate India Private Limited is seeking a senior engineer to design and build AI-driven data platforms. You will develop agentic AI pipelines, create enterprise ontologies, and deliver LLM-powered services for semantic enrichment and metadata generation.

The role emphasizes scalable Python-based microservices, RDF/OWL ontologies, SPARQL reasoning, and cloud-native data pipelines on Azure/Microsoft Fabric.

Qualifications

• 6+ years of professional software engineering experience.
• Strong proficiency in Python and GenAI.
• Hands-on experience building LLM-based systems.

Responsibilities

• Design, build, and maintain agentic AI pipelines.
• Create and evolve enterprise ontologies in RDF/OWL.
• Engineer LLM-powered services for schema understanding and metadata generation.
• Implement SPARQL querying and reasoning over knowledge graphs.
• Architect and deliver Python-based microservices and batch pipelines.
• Build and optimize dimension and fact generation pipelines on Microsoft Fabric.
• Define and enforce engineering standards and reusable components for semantic platforms.
• Partner with data architects and governance teams to validate semantic definitions.
• Conduct code reviews and mentor engineers to raise quality.

Skills

Python

GenAI

LLM systems

RDF/OWL

SPARQL

Data engineering

Azure / Fabric

Google ADK

LangChain

ETL/ELT

Knowledge graphs

Tools

Google ADK

LangChain

Microsoft Fabric

Azure

Spark

Job description

Job Description

Key Responsibilities

• Design, build, and maintain agentic AI pipelines (using Google ADK or similar frameworks) to automate semantic mapping, dimension mining, and ontologydriven reasoning.
• Create and evolve enterprise ontologies in RDF/OWL, including upper ontologies and domain extensions aligned to CIM (where applicable), to enable reusable enterprise semantics.
• Engineer LLMpowered services for schema understanding, semantic alignment, ontology enrichment, and AIassisted metadata generation, with a focus on accuracy, traceability, and scale.
• Implement SPARQL querying and reasoning layers over knowledge graphs to drive downstream transformations and ensure consistent interpretation of business concepts.
• Architect and deliver Pythonbased microservices and batch pipelines that integrate semantic reasoning with modern dataengineering workflows.
• Build and optimize dimension and fact generation pipelines on Microsoft Fabric (Lakehouse, Spark, SQL, orchestration) to produce businessready star schemas from heterogeneous sources.
• Define and enforce engineering standards, design patterns, and reusable components for semantic and AIdriven data platforms (quality, observability, security, and performance).
• Partner with data architects, domain SMEs, and governance teams to validate semantic definitions, manage change, and ensure platform scalability and adoption.
• Conduct code reviews, mentor engineers, and influence technical decisions across the platform to raise engineering quality and delivery velocity.

Required Skills & Qualifications:

• 6+ years of professional software engineering experience, with strong proficiency in Python and GenAI.
• Handson experience building LLMbased systems using commercial or opensource models.
• Solid understanding of semantic technologies: RDF, OWL, ontologies, knowledge graphs, and SPARQL.
• Experience designing or working with agentic AI frameworks (e.g., Google ADK, LangChain agents, or similar).
• Strong background in data engineering concepts (ETL/ELT, star schemas, metadatadriven pipelines).
• Experience building and operating systems on cloud platforms, preferably Microsoft Azure / Microsoft Fabric.
• Strong problemsolving skills and ability to work in ambiguous, greenfield platform initiatives.

Preferred Skills:

• Experience with enterprise data models (e.g., CIM or canonical models).
• Familiarity with semantic alignment, ontology mapping, or data cataloguing tools.
• Exposure to MLOps / LLMOps, model evaluation, and AI observability.
• Knowledge of distributed systems, CI/CD pipelines, and containerisation.
• Experience building AIassisted analytics or semantic layers for BI or NLQ use cases.