about 4 hours ago
Bengaluru, IndiaMid Level / Senior
Responsibilities
- Build and maintain components of the agentic runtime for multi-turn financial conversations.
- Implement context management logic for tracking consumer state and conversation history.
- Write and iterate on primitives balancing conversational fluidity with structured reasoning.
- Work on real-time voice pipeline aiming for sub-1s latency.
- Contribute to VAD tuning and turn-taking logic for natural conversations.
- Evaluate and integrate speech models for various applications.
- Debug streaming audio and WebRTC issues in production.
- Write and refine prompts and implement orchestration flows.
- Build components of the evaluation framework for agent quality measurement.
Requirements
- 2 – 5 years of hands-on engineering experience with ML or AI systems.
- Solid Python fundamentals and ability to write clean, maintainable code.
- Some experience with LLMs, including prompt engineering and API integrations.
- High bias for action and ability to ship work without over-engineering.
- Familiarity with TypeScript is a plus.
- Eager to learn and take ownership in a fast-moving environment.
Benefits
- Real ownership from day one in a small pod.
- Opportunity to work on frontier voice AI problems.
- Strong mentorship from senior engineers and AI Lead.
- High leverage early career role impacting millions of financial conversations.