Role & responsibilities

Voice AI Development

- Design and build real-time voice-based AI agents capable of natural, low-latency conversations.

- Develop and integrate Automatic Speech Recognition (ASR), Text-to-Speech (TTS), and Natural Language Understanding (NLU) pipelines entirely or predominantly in-house.

- Implement wake-word detection, speaker diarization, noise cancellation, and voice activity detection (VAD) modules.

- Build dialogue management systems with context tracking, intent handling, and multi-turn conversation flows.

- Optimize ASR/TTS models for accuracy, latency, and domain-specific vocabulary.

Infrastructure & Platform Engineering

- Design and manage the cloud or on-premise infrastructure required to deploy, scale, and monitor voice AI workloads.

- Set up CI/CD pipelines, containerization (Docker/Kubernetes), and model serving infrastructure (e.g., Triton, TorchServe, custom gRPC services).

- Ensure high availability, fault tolerance, and low-latency performance of real-time voice systems.

- Implement logging, monitoring, and alerting for production AI systems.

- Reduce dependency on external vendors by building reusable, internal libraries and services.

In-House Solution Design

- Evaluate open-source ASR (e.g., Whisper, Wav2Vec2, Kaldi), TTS (e.g., Coqui, VITS, FastSpeech2), and LLM frameworks and adapt them for production use.

- Build custom model training and fine-tuning pipelines for domain-specific voice applications.

- Contribute to internal tooling, SDKs, and APIs that allow other teams to build voice-powered features.

Collaboration & Leadership

- Collaborate with product, design, and data teams to translate business requirements into technical solutions.

- Conduct code reviews, mentor junior engineers, and enforce engineering best practices.

- Document architecture decisions, APIs, and operational runbooks.

Required Qualifications

Experience





- 5 years of total software development experience.

- Minimum 1 year of hands-on experience building and deploying AI agents in a production environment.

- Proven track record of building in-house AI / ML solutions with limited dependency on external SaaS platforms.

Technical Skills Voice AI

- Strong experience with ASR systems: Whisper, Wav2Vec2, DeepSpeech, Kaldi, or equivalent.

- Experience with TTS engines: Coqui TTS, VITS, FastSpeech2, Mozilla TTS, or equivalent.

- Hands-on experience with NLU/dialogue frameworks: Rasa, custom transformer-based models, or similar.

- Familiarity with audio processing libraries: librosa, PyDub, SpeechBrain, WebRTC VAD.

- Understanding of real-time audio streaming protocols (WebSockets, WebRTC, SIP/VoIP).

Technical Skills – Infrastructure

- Proficiency with Python and worked around 1 year in PHP-Larvel.

- Experience with Docker, Kubernetes, and cloud platforms (AWS, GCP, or Azure).

- Familiarity with ML model serving and optimization (ONNX, TensorRT, Triton Inference Server).

- Experience with message brokers and streaming platforms (Kafka, RabbitMQ, or Redis Streams).

- Working knowledge of databases: PostgreSQL, Redis, and vector databases (Pinecone, Weaviate, or Qdrant).

Good to Have

- Experience with on-device / edge AI for voice (TFLite, CoreML, ONNX Runtime).

- Familiarity with telephony stacks and VoIP platforms (Asterisk, FreeSWITCH, Twilio Programmable Voice).

- Experience fine-tuning LLMs (LLaMA, Mistral, Falcon) for conversational use cases.

- Contributions to open-source voice AI or ML projects.

What We Offer

- Chance to build a cutting-edge voice AI platform from the ground up.

- High ownership, low bureaucracy — your decisions directly shape the product.

- Competitive compensation and performance-linked incentives.

- Flexible working arrangements.

- Access to GPU compute resources and a dedicated AI research budget.

📌 Technical Project Manager (Dehradun)
🏢 BigShip
📍 Dehradun