Full Stack AI Engineer Company: Namasys Analytics (Kynaa AI)

Location: India (remote-friendly)

Type: Full-time

Experience: 4 to 7 yrs About the role

We are building a real-time, voice-driven AI product that holds live spoken conversations with users and produces AI-scored assessments. The centre of gravity is backend and AI. A real-time voice agent that conducts live conversations, an AI scoring pipeline that turns transcripts into assessment reports, and a family of LLM-powered features around documents and content generation.

This is not a role where AI is a garnish; a lot of the time you will be working directly with model behaviour, prompts, token economics and evaluation. What you will do

Build and extend LLM-powered features end to end: prompt design, structured-output calls, retry and failure handling, cost and token accounting, and the evaluation harnesses that prove a change is safe before it ships.

Grow into the realtime voice agent: our conversational engine runs on a realtime speech-to-speech model over WebRTC. You are not expected to arrive with this experience; you are expected to make it yours within months, working alongside the founder.

Own backend services: Python/FastAPI, async SQLAlchemy, background workers, queues and the API surface that serves the product's web applications.

Work across the stack when a feature needs it: our frontend is Next.js/React/TypeScript. You will wire APIs into existing UI and modify components confidently.

Leave the system more understandable than you found it:



decisions here are written down and binding, quality gates are non-negotiable. What we require

4 to 7 years building production software, with solid Python and real depth in at least one backend framework (we run FastAPI).

Shipped LLM-integrated features in production. You have called model APIs in real products, dealt with non-determinism, structured outputs, prompt iteration and cost, and can talk concretely about what went wrong and how you found out.

Solid API and data fundamentals: SQL, async patterns, queues or background processing.

Enough React/TypeScript to modify components and wire interfaces without hand-holding.

Comfort working in a small team with high ownership, written decision-making and a strong code review culture.

Clear written and spoken English; the product itself is conversational and you will regularly use it aloud to test your own work. What we would love

Realtime or voice experience: WebRTC, LiveKit, telephony, speech models.

Experience with LLM evaluation: golden sets, A/B harnesses, regression baselines.

Azure familiarity (we run on Container Apps, Azure SQL and Service Bus).

Prior work in an early-stage or very small team where you shipped whole features alone. How we work

Small team, high trust, direct communication, modern AI-assisted engineering tooling. Production is actively developed and ships frequently behind automated quality gates.

Interested

Candidates can share their resume at [email protected]

📌 Full Stack AI Engineer (India)
🏢 Namasys Analytics
📍 India