AI ML Engineer

About Hero Vired:

Would you like to be part of an exciting, innovative, and high-growth startup from one of the largest and most well-respected business houses in the country - the Hero Group?

Hero Vired is a premium learning experience offering industry-relevant programs and world-class partnerships, to create the change-makers of tomorrow.

At Hero Vired, we believe everyone is made of big things. With the experience, knowledge, and expertise of the Hero Group, Hero Vired is on a mission to change the way we learn. Hero Vired aims to give learners the knowledge, skills, and expertise through deeply engaged and holistic experiences, closely mapped with industry to empower them to transform their aspirations into reality. The focus will be on disrupting and reimagining university education & skilling for working professionals by offering high-impact online certification and degree programs.

The illustrious and renowned US$5 billion diversified Hero Group is a conglomerate of Indian companies with primary interests and operations in automotive manufacturing, financing, renewable energy, electronics manufacturing, and education. The Hero Group (BML Munjal family) companies include Hero MotoCorp, Hero FinCorp, Hero Future Energies, Rockman Industries, Hero Electronix, Hero Mindmine, and the BML Munjal University.

For detailed information, visit Hero Vired

Role : AI ML Engineer

Job Type: Full Time

Work Tyoe: Work From Office

Location: New Delhi (Sultanpur)/Gurugram

Experience: 3+ years

About the Role:

Hero Vired is seeking a skilled AI ML Engineer Conversational AI Systems with 3+ years of relevant experience to support the design, development, and optimization of intelligent conversational platforms used in our digital learning ecosystem. This role will focus on building scalable data pipelines, training-ready datasets, and AI-driven conversational solutions that enhance learner engagement and personalized experiences.

The ideal candidate will work closely with AI/ML engineers, product teams, and content specialists to develop robust data foundations for chatbots, virtual teaching assistants, and AI-powered learner support systems. You will play a key role in enabling NLP models, managing structured and unstructured data, and ensuring high-quality, reliable data flows for conversational AI applications.

This position is ideal for professionals passionate about AI, NLP, and education technology, who enjoy working in fast-paced environments and want to create real-world impact through AI-driven learning solution.

Key Responsibilities

Conversational AI Infrastructure

• Build and optimize knowledge pipelines for voice bot context injection using Azure AI Search + vector embeddings
• Design ETL pipelines to process CRM data (lead profiles, interaction history, program details) for real-time bot personalization
• Implement lazy-recall architecture for efficient prompt context loading during live calls

Voice Bot Data Layer

• Structure product/program data (courses, pricing, features, eligibility) in normalized formats optimized for sub-100ms retrieval
• Build semantic search pipelines for dynamic FAQ retrieval and objection handling content
• Design customer persona classification models to enable adaptive conversation flows

Real-Time Processing & Integration

• Develop data sync pipelines between CRM, LMS (Moodle), and bot orchestration layer
• Build streaming pipelines for call metadata, conversation state, and lead scoring updates
• Integrate with Azure services (Cognitive Services, OpenAI, Cosmos DB) for knowledge base management

Quality & Feedback Systems

• Design conversation analytics pipelines to track bot performance metrics (booking rates, objection handling success, human-likeness scores)
• Implement LLM-based quality analysis for automated transcript evaluation
• Build feedback loops that feed conversation insights back into prompt optimization and knowledge base updates

ML & Continuous Improvement

• Train/update classification models for intent detection, customer sentiment, and lead scoring
• Implement confidence scoring and hallucination detection for bot responses
• Design RLHF pipelines using conversation quality scores to fine-tune response generation
• Build reward models that learn from QA analyst feedback and successful conversion patterns
• Implement RAG pipelines for dynamic knowledge retrieval during live calls

Deployment & MLOps (End-to-End Ownership):

• Own CI/CD pipelines for prompt versioning, knowledge base updates, and model deployments
• Design multi-environment strategy (dev staging prod) with automated testing for bot responses
• Implement blue-green/canary deployments for safe rollout of prompt changes and model updates
• Build automated regression testing validate bot responses against golden datasets before production push
• Set up infrastructure-as-code (Terraform/Bicep) for Azure resources
• Design rollback mechanisms for quick recovery when bot performance degrades
• Implement observability stack: logging, tracing, alerting for latency spikes, error rates, and conversation quality drops
• Manage containerized deployments (Docker/Kubernetes) for bot services and data pipelines
• Automate knowledge base refresh cycles with validation gates

Required Core Skill Sets

• Python (FastAPI, Pandas, Scrapy) + PostgreSQL/Cosmos DB
• Azure AI stack (AI Search, OpenAI, Cognitive Services, Azure Functions)
• RAG architecture and vector embeddings
• Real-time data pipelines (Airflow/Prefect)

MLOps & DevOps

• CI/CD tools (Azure DevOps, GitHub Actions, or GitLab CI)
• Infrastructure-as-code (Terraform/Bicep)
• Docker, Kubernetes (AKS preferred)
• Monitoring & observability (Prometheus, Grafana, Azure Monitor)
• Git-based prompt/config versioning strategies

Preferred

• Experience with conversational AI or voice systems
• RLHF or preference-based model tuning
• A/B testing frameworks for ML systems