- Lead implementation of LLM-based features summarization, sentiment detection, auto-disposition, escalation tagging.

- Fine-tune and evaluate models (Whisper, GPT, HuggingFace, Rasa) for vernacular (Indian) language support.

- Build and deploy LangChain pipelines for prompt engineering, QA tagging, and agent assist.

- Prototype emotion recognition, contextual agent replies, and a real-time assist layer.

- Build and maintain inference pipelines using FastAPI, Docker, and Kubernetes.

- Integrate AI modules into core product features (Dialer, CRM sync, IVR).

- Optimize model latency and deployment strategy for high-concurrency environments.

- Architect scalable data pipelines using PostgreSQL, Redis, and Kafka.

- Build ETL/ELT workflows to support real-time analytics, dashboards, and feedback loops.

- Maintain secure, compliant data storage, retrieval, and access control pipelines (DPDP, GDPR-ready).

- Work closely with Product, Engineering, and UX to deliver features that impact agent productivity.

- Guide junior ML and data engineers; define and enforce coding/data standards.

- Contribute to AI strategy, model governance, and data infrastructure roadmap.

Requirements :

- 6+ years of overall software engineering experience, with 3+ years hands-on in AI, ML, Generative AI, or Applied AI engineering.

- Strong hands-on experience designing, developing, and deploying AI applications using LLMs, RAG, Agentic AI frameworks, and Generative AI architectures.

- 1+ recent year of hands-on experience in Voice Bot Development, Voice AI, Conversational AI, AI Voice Agents, Speech AI, Contact Center Automation, or Voice Automation Platforms.

- Strong Python expertise built scalable AI/ML applications, APIs, microservices, or backend systems using Python-based frameworks.

- Hands-on experience with AI/LLM frameworks such as LangChain, LangGraph, HuggingFace, LlamaIndex, CrewAI, AutoGen, Whisper, OpenAI SDKs, or equivalent GenAI frameworks.

Good to Have :

- Experience with AWS, Azure, GCP, MLOps, AI deployment platforms, model serving infrastructure, and cloud-native architectures.