Applied Scientist - Agentic AI

Responsibilities:

• Build and ship chat, voice, and multimodal AI agents integrated into production products.
• Design agent workflows using LLMs, tools, memory, and state.
• Implement voice pipelines, including ASR, TTS, turn-taking, and interruptions.
• Work on Indian language voice agents, including latency and quality tuning.
• Develop workflow automation logic driven by agents.
• Build task execution systems where agents plan, act, and verify outcomes.
• Design and implement media and martech agents as part of product features.
• Build persona-based systems for content, campaigns, and engagement.
• Implement agent moderation and control mechanisms.
• Add observability, logging, and debugging for AI behaviour.
• Reduce hallucinations and failure modes through system design.
• Collaborate closely with product, design, and engineering teams.

Requirements:

• Master's or PhD in Computer Science, Artificial Intelligence, Machine Learning, or Data Science.
• Hands-on experience building LLM-powered applications.
• Experience with chat agents or conversational systems.
• Understanding of tool calling, agent loops, memory, and state.
• Experience integrating AI, APIs and MCPs into backend or product workflows.
• Familiarity with voice AI systems or strong interest in building them.
• Ability to rapidly prototype and iterate on AI-driven features.
• Strong debugging mindset for real-world AI failures.
• Comfort working in fast-moving product environments.

Strong plus if you have experience with:

• Voice agents, speech-to-text, text-to-speech, telephony systems, Indian language NLP or speech systems.
• Workflow automation or agent orchestration frameworks.
• RAG systems, embeddings, and vector databases.
• Martech, CLM, campaign automation, or ad-tech workflows.
• Persona-based content or personalisation systems.
• SaaS products with real users and scale constraints.