Senior AI/ML Engineer II
Sign up free to see how well your resume matches this role.
What you'll do
- Work on our real-time voice pipeline (ASR → LLM → TTS) built on Pipecat, integrating providers like ElevenLabs, Deepgram, OpenAI, Azure, and Cartesia (plus our in-house speech models)
- Latency: making the agent respond in well under a second — turn detection, speculative generation, model warm-up, filler/latency-hiding techniques, and profiling
- Conversation quality: turn-taking, interruption/barge-in, silence and voicemail handling, DTMF, and multi-step conversation flows
- LLM orchestration: prompting, tool/function calling, and outcome classification
- RAG: grounding answers in knowledge bases (Azure AI Search / Qdrant)
- Evaluation: extending our offline eval harness (incl. LLM-as-judge) to measure and improve ASR accuracy, response quality, and TTS naturalness
- Reliability and scale: error handling, fallbacks, and observability (OpenTelemetry/Azure Monitor) for high call volumes across regions
Summarised by NextRaise from the employer’s description, which follows in full below.
Full description from employer
About Chubb
Chubb is a world leader in insurance. With operations in 54 countries and territories, Chubb provides commercial and personal property and casualty insurance, personal accident and supplemental health insurance, reinsurance and life insurance to a diverse group of clients. The company is defined by its extensive product and service offerings, broad distribution capabilities, exceptional financial strength and local operations globally. Parent company Chubb Limited is listed on the New York Stock Exchange (NYSE: CB) and is a component of the S&P 500 index. Chubb employs approximately 43,000 people worldwide. Additional information can be found at: www.chubb.com.
About Chubb India
At Chubb India, we are on an exciting journey of digital transformation driven by a commitment to engineering excellence and analytics. We are proud to share that we have been officially certified as a Great Place to Work® for the third consecutive year, a reflection of the culture at Chubb where we believe in fostering an environment where everyone can thrive, innovate, and grow.
With a team of over 2500 talented professionals, we encourage a start-up mindset that promotes collaboration, diverse perspectives, and a solution-driven attitude. We are dedicated to building expertise in engineering, analytics, and automation, empowering our teams to excel in a dynamic digital landscape.
We offer an environment where you will be part of an organization that is dedicated to solving real-world challenges in the insurance industry. Together, we will work to shape the future through innovation and continuous learning.
Position Details:
- Function/Department: Advanced Analytics
- Location: Bangalore, India
- Employment Type: Full-time
- Role Overview – Senior AI/ML Engineer II
As a Senior AI Engineer for Voice AI, you will architect, develop, and optimize advanced real-time conversational AI systems for call center environments. You will lead technical initiatives, mentor junior engineers, and ensure the delivery of robust, scalable, and ultra-low latency voice-to-voice solutions using speech recognition, large language models (LLMs), and text-to-speech (TTS) technologies
Responsibilities
- Work on our real-time voice pipeline (ASR → LLM → TTS) built on Pipecat, integrating providers like ElevenLabs, Deepgram, OpenAI, Azure, and Cartesia (plus our in-house speech models)
- Latency: making the agent respond in well under a second — turn detection, speculative generation, model warm-up, filler/latency-hiding techniques, and profiling
- Conversation quality: turn-taking, interruption/barge-in, silence and voicemail handling, DTMF, and multi-step conversation flows
- LLM orchestration: prompting, tool/function calling, and outcome classification
- RAG: grounding answers in knowledge bases (Azure AI Search / Qdrant)
- Evaluation: extending our offline eval harness (incl. LLM-as-judge) to measure and improve ASR accuracy, response quality, and TTS naturalness
- Reliability and scale: error handling, fallbacks, and observability (OpenTelemetry/Azure Monitor) for high call volumes across regions
Company
Company facts come from this company's own listings. We only show what the postings themselves carry.