New opportunity

Data Back End Developer - Python with AI

Vontier · Bangalore, India

About this role

Driivz, a Vontier company, is seeking a Data Back End Developer to join their team focused on the e-mobility revolution. The role involves designing and building agentic services and AI-augmented platform capabilities, including context and harness engineering, and tool/MCP integrations to connect LLMs to core systems. You will own reliability end-to-end by building evaluation suites, guardrails, and observability to ensure systems are production-grade. Additionally, you will enhance core capabilities in LOB areas such as Billing and CRM through AI-augmented workflows.

Responsibilities

Design and build agentic services and AI-augmented platform capabilities.
Develop and maintain evaluation suites, guardrails, and observability for AI systems.
Enhance core LOB capabilities including Billing and CRM.
Collaborate within a team to build scalable, integrated SaaS solutions.

Required Qualifications

Minimum 5 years of proven Python experience.
Proven AI Engineering experience (Context/Harness engineering, Tools, MCP, etc.).
Experience making agentic/LLM systems reliable in production.
Proven experience with Microservices Architecture.
Proven experience in SaaS multitenant high throughput systems.
Proven experience with Relational Databases (MySQL/Postgres, etc.).
Experience with Docker.

Preferred Qualifications

Experience with agent orchestration frameworks (LangChain, LangGraph).
Familiarity with Python AI/web frameworks (FastAPI, FastMCP, Pydantic).
Experience with retrieval and vector search (pgvector).
Familiarity with LLM observability and evaluation tooling (OpenTelemetry, Langfuse).
Experience with Kubernetes, public cloud (AWS preferred), event-driven architecture (Kafka), and big-data/analytics warehouses (Snowflake, Redshift).
Knowledge of Java/Kotlin (Spring Boot) for integration with JVM platforms.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonAI EngineeringAgentic ServicesLLMMCPMicroservices ArchitectureSaaSMySQLPostgresDockerLangChainLangGraphFastAPIFastMCPPydanticpgvectorOpenTelemetryLangfuseKubernetesAWSKafkaSnowflakeRedshiftJavaKotlinSpring Boot

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Production-Grade AI Reliability
The role requires moving beyond AI prototypes to building robust evaluation suites, guardrails, and observability frameworks. You must demonstrate how to measure and maintain the performance of agentic services in a high-throughput SaaS environment.
Agentic Service Integration
You will design and build agentic services that connect LLMs to core business systems using context engineering, harness engineering, and tool/MCP integrations. Success depends on your ability to bridge the gap between AI models and existing enterprise workflows.
Scalable SaaS Architecture
As part of a multitenant platform, you must apply microservices architecture principles to ensure AI-augmented workflows in Billing and CRM remain performant and scalable under heavy load.

Likely challenges to think through

Integrating AI with Legacy JVM Platforms
The role involves integrating Python-based AI services with an existing Java/Kotlin (Spring Boot) platform. Prepare to discuss strategies for cross-language communication, data consistency, and maintaining system integrity during these integrations.
Balancing Frameworks vs. Primitives
While experience with LangChain or LangGraph is preferred, the role values the ability to build directly on LLM primitives when necessary. Be ready to justify when to use an abstraction layer versus a custom, lightweight implementation to optimize for performance and control.

Practice your answers

Technical practice

Designing Agentic Services in Python

What to coverExplain your approach to building agentic services that interact with core systems. Focus on how you handle context engineering, tool selection, and the integration of Model Context Protocol (MCP) to ensure LLMs perform reliable tasks within a production environment.

Ensuring Reliability in LLM Systems

What to coverDescribe the strategies you use to move AI systems from prototypes to production-grade services. Discuss your experience with implementing evaluation suites, guardrails, and observability tools like Langfuse or OpenTelemetry to monitor performance and accuracy.

Microservices and Multitenant SaaS Architecture

What to coverHow do you design high-throughput microservices for a multitenant SaaS platform? Address challenges related to data isolation, scalability, and maintaining consistent performance across different tenants in a distributed architecture.

Database Optimization for AI Workflows

What to coverDiscuss your experience with relational databases like PostgreSQL in the context of AI applications. Specifically, explain how you utilize vector search capabilities, such as pgvector, to support retrieval-augmented generation (RAG) workflows.

Event-Driven Architecture with Kafka

What to coverExplain how you integrate event-driven patterns using Kafka into a backend system. Describe how you handle data streaming or change data capture (CDC) to keep AI-augmented services synchronized with core LOB systems like Billing or CRM.

Orchestration Frameworks vs. Custom Primitives

What to coverCompare the trade-offs between using agent orchestration frameworks like LangChain or LangGraph versus building directly on raw LLM primitives. When would you choose a lightweight custom implementation over a heavy framework?

Containerization and Orchestration

What to coverDescribe your workflow for deploying Python-based AI services using Docker and Kubernetes. Focus on how you manage resource allocation, scaling, and environment consistency for services that may have varying compute requirements.

Integrating Python Services with JVM Platforms

What to coverGiven the need to integrate with Java/Kotlin Spring Boot platforms, how do you handle cross-language communication and data consistency? Discuss your approach to API design and shared service contracts.

Behavioural practice

Handling Production Failures in AI Systems

What to coverDescribe a situation where an automated system or AI model failed in production. Explain the steps you took to identify the root cause, implement guardrails to prevent recurrence, and communicate the impact to stakeholders.

Collaborating Across Technical Domains

What to coverProvide an example of working within a cross-functional team to integrate a new AI feature into an existing legacy system. How did you manage technical debt while ensuring the new solution met performance requirements?

Balancing Speed and Quality in AI Development

What to coverTell me about a time you had to balance the need for rapid prototyping of an AI agent with the requirement for long-term reliability. How did you decide which evaluation metrics were critical to prioritize?

Adapting to Rapidly Evolving AI Technologies

What to coverThe AI landscape changes quickly; describe a time you had to learn and implement a new tool or framework (like a new MCP integration or observability tool) under a tight deadline. How did you ensure the implementation was robust?

Sharpen your knowledge

Topics to prepare

Agentic System Reliability
Focus on the conceptual framework of building production-grade LLM systems. Exercise: Design a strategy for implementing guardrails and evaluation suites to detect hallucinations or latency issues in an agentic workflow. This is critical for the role's requirement to move beyond demos to production-grade AI.
Microservices and Event-Driven Architecture
Review patterns for high-throughput, multitenant SaaS platforms. Exercise: Map out how an AI-augmented service would interact with existing Billing or CRM microservices using event-driven patterns like Kafka. This aligns with the need to integrate AI into core platform capabilities.

Skills to revise

Python AI Engineering (MCP and Tooling)
Practice connecting LLMs to external systems using Model Context Protocol (MCP). Exercise: Build a small Python service using FastMCP that exposes a database tool to an LLM agent. This directly addresses the requirement for context engineering and tool integration.
LLM Observability and Evaluation
Gain hands-on experience with tools like Langfuse or OpenTelemetry. Exercise: Set up a basic trace for an LLM call to monitor latency and token usage, and define a simple automated evaluation metric for response accuracy. This is essential for the 'own reliability end-to-end' responsibility.
Vector Search Implementation
Practice retrieval-augmented generation (RAG) using pgvector. Exercise: Create a schema in a Postgres database with a vector column and perform a similarity search on a sample dataset. This is highly relevant to the preferred qualifications for retrieval and vector search.

Your preparation checklist

Build a production-grade LLM evaluation suite
Create a project that demonstrates how you measure LLM performance. Include guardrails and observability metrics using tools like Langfuse or OpenTelemetry to show you can move beyond simple demos to reliable, production-ready agentic services.
Develop an agentic service with tool integration
Prepare a code walkthrough of a Python-based agent that utilizes MCP or similar tool-calling frameworks to interact with a core system. Focus on how you handle context engineering and ensure the agent performs reliably within a microservices architecture.
Review SaaS multitenancy and high-throughput patterns
Be ready to discuss your experience designing scalable, multitenant SaaS systems. Prepare examples of how you have managed high-throughput data flows using relational databases like PostgreSQL and event-driven architectures like Kafka.
Prepare for JVM and Python integration discussions
Since the role involves integrating with a JVM-based platform, prepare to explain how you would bridge Python-based AI services with Java/Kotlin Spring Boot environments, highlighting your experience with API design and cross-language interoperability.
Formulate technical questions for the team
Prepare thoughtful questions regarding their current AI observability stack, how they manage guardrails for billing-related workflows, and the specific challenges they face when scaling agentic services within their EV charging platform.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…