Website:
aivar.tech
Job details:
About Aivar Innovations
Aivar Innovations is an AI-native services and software company and an AWS Preferred Partner, backed by Bessemer Venture Partners and Sorin Investments. We ship production-grade AI into enterprise environments across fintech, healthcare, and technology.
Our work is anchored by four accelerator platforms:
Convogent (voice and agent AI automation),
Velogent (governed agentic process automation for regulated industries),
Kubogent (Kubernetes-native AIOps), and
Datagent (data and analytics). We measure ourselves by what runs in production — not by slideware.
Role Overview
We are hiring a
Conversational AI Engineer to architect and ship voice AI agents that handle thousands of live customer conversations daily. You will own real-time audio pipelines, telephony integrations, and the orchestration layer that ties STT, TTS, and LLMs into reliable sub-second voice experiences. This is a hands-on engineering role at the core of our Convogent accelerator.
What You'll Do
- Build voice agents. Design and deploy scalable voice AI agents using Pipecat, LiveKit, or equivalent frameworks.
- Engineer real-time audio pipelines. Optimize STT and TTS flows for low latency, high concurrency, and call quality.
- Own telephony integration. Implement WebRTC, SIP, RTP, and WebSocket connectivity, and integrate with CPaaS platforms like Twilio and Telnyx for inbound and outbound calling.
- Handle multi-speaker audio. Build speaker diarization and segmentation for accurate, context-aware conversations.
- Build the orchestration layer. Connect STT, TTS, and LLM agents into resilient, observable workflows with guardrails and evaluation hooks.
- Tune for production. Implement Voice Activity Detection, echo cancellation, codec optimization, and call quality monitoring.
- Operate at scale. Monitor, log, and support production deployments on AWS — including incident response and latency tuning.
- Partner with product and customer teams. Translate customer requirements into deployable voice solutions and iterate against measurable outcomes.
What You'll Bring
- Hands-on experience building and deploying voice agents with Pipecat, LiveKit, or similar frameworks.
- Deep expertise in STT (Deepgram, Whisper, AssemblyAI, Google STT, Speech) and TTS (ElevenLabs, Cartesia, Amazon Polly, TTS).
- Strong working knowledge of telephony, SIP, WebRTC, and media servers (FreeSWITCH, Asterisk, Kamailio).
- Experience with multi-speaker diarization and conversation segmentation.
- Production fluency in Python plus at least one of Node.js, TypeScript, or Go.
- Practical use of AI/ML libraries (TensorFlow, PyTorch, Scikit-learn) and conversational AI design — prompt engineering and integration with LLMs (GPT-4o, Claude, Bedrock models).
- Audio codec depth (Opus, G.711, G.729), real-time streaming, and media handling.
- Cloud deployment experience on AWS (Lambda, EC2, S3, EKS, Polly, Transcribe, Bedrock) with Docker, Kubernetes, and CI/CD.
- Demonstrated track record of shipping voice AI or telephony systems to production with measurable customer impact — latency targets met, call volume handled, quality metrics moved.
Preferred Experience
- Experience with media server architectures (SFU/MCU) and custom VoIP infrastructure (Asterisk, FreeSWITCH, Kamailio, OpenSIPS).
- Familiarity with low-code voice platforms alongside custom infrastructure.
- Voice analytics, transcription tooling, and Indian-language STT/TTS models.
- AWS or speech technology certifications.
- Experience leading cross-functional delivery and working with enterprise stakeholders.
Why This Role, Why Aivar
- Work that ships. Your voice agents will run live on enterprise calls — measured by customer outcomes, not demos.
- Backed and proven. AI-native, AWS Preferred Partner, BVP- and Sorin-backed, with 100+ enterprise customers in the first year.
- Own the stack. End-to-end ownership from real-time audio to LLM orchestration to production operations. No layers between you and the system.
- Build on accelerators that work. Convogent has real production traction — you extend a working platform, not start from zero.
- Shape the roadmap. Direct collaboration with founders, product, and the AI engineering bench. Your field signal drives what gets built next.
- Compensation. Competitive base, performance bonus, ESOPs, AWS/AI learning budget, and conference sponsorship.
Click on Apply to know more.