Generative AI Engineer

Job Description: Senior AI Engineer – VoizPanda (AI Multi-Calling Platform)

Location: BasaveshwaraNagar,

Bangalore

Mode: Onsite (Full-time)

Experience: 2–7 years

Compensation: ₹6–9 LPA

About VoizPanda

VoizPanda is Digi9’s in-house AI-powered multi-calling platform that enables enterprises to run thousands of simultaneous voice calls handled entirely by intelligent conversational AI agents.

These agents can listen, understand, and respond naturally — creating human-like phone interactions for lead generation, customer service, and data collection.

Role Overview

The Senior AI Engineer will build and optimize the intelligence layer that powers VoizPanda’s real-time AI conversations.

You’ll work across speech recognition, intent understanding, dialogue management, and speech synthesis, ensuring ultra-low latency and human-like interaction quality.

You’ll collaborate with product, backend, and DevOps teams to train, deploy, and scale models that serve live production workloads across industries such as banking, solar, education, and healthcare.

Key Responsibilities

- Design, train, and fine-tune ML models for speech recognition (ASR) and intent classification.
- Integrate and optimize LLM/NLU pipelines (AI4Bharat, OpenAI, Hugging Face, Rasa, LangChain).
- Build real-time inference pipelines for low-latency conversational flow.
- Develop and monitor evaluation frameworks for latency, accuracy, and dialogue coherence.
- Collaborate with backend engineers to integrate AI microservices using REST or GraphQL APIs.
- Manage MLOps workflows — containerization (Docker/Kubernetes), model versioning, and deployment.
- Monitor and retrain production models to handle drift, new intents, or accents.
- Maintain data pipelines for training, annotation, and validation datasets.

Must-Have Skills

- Python (advanced), TensorFlow or PyTorch, and Hugging Face Transformers
- Speech/NLP experience with Whisper, wav2vec, or Rasa
- Proficiency in NLU/NLP frameworks (LangChain, spaCy)
- MLOps stack: Docker, Kubernetes, MLflow, FastAPI
- Understanding of SaaS-based AI systems and real-time pipelines
- Familiarity with REST/GraphQL microservices and API-based AI integrations

Good-to-Have Skills

- Vector databases (FAISS, Pinecone, Weaviate)
- GPU optimization for inference (TensorRT, ONNX Runtime)
- Voice synthesis / text-to-speech (TTS) or voice cloning projects
- Experience testing AI models for bias, drift, or latency