Kapture CX - Technical Product Manager - Applied AI

Description :

This is a hands-on, high-impact role focused on improving the real-world performance of our Voice AI platform.

You will work at the intersection of LLMs, voice systems, and product behavior testing new models, experimenting with prompting strategies, evaluating configurations and identifying what actually works in production.

You will not just support implementations you will shape how the product evolves by feeding structured insights from experimentation, debugging and real-world usage back into the product roadmap.

What will you do ?

Model Evaluation & Experimentation :

- Test and benchmark different LLMs, STT/TTS systems and configurations across use cases.

- Evaluate performance across latency, accuracy, cost and conversation quality.

- Run structured experiments to compare prompts, multi-agent setups and orchestration strategies.

Voicebot Behavior & Prompting :

- Design and refine prompts, guardrails, fallback strategies and conversation flows.

- Improve handling of noisy ASR, multilingual inputs and edge-case scenarios.

- Continuously optimize conversation quality based on real call data.

System Debugging & Root Cause Analysis :

- Analyze logs, traces and transcripts to identify failure points.

- Distinguish between model issues, prompt gaps, platform limitations and integration

problems.

- Propose and validate fixes through rapid iteration.

Product Feedback & Roadmap Influence :

- Identify recurring gaps and translate them into clear product requirements.

- Work closely with Product and Engineering teams to prioritize improvements.

- Act as the voice of what works in production for the product roadmap.

Applied Problem Solving :

- Develop practical workarounds for current system limitations.

- Balance ideal AI behavior with real-world constraints (latency, APIs, data availability).

- Improve system reliability and robustness across deployments.

What does success look like in this role ?

- Measurable improvement in voicebot accuracy, latency, and conversation quality.

- Faster identification and resolution of system issues.

- Clear, actionable inputs into product and model decisions.

- Better adoption of new models and architectures (e.g., multi-agent setups).

- Strong alignment between real-world performance and product capabilities.

What would make you a good fit ?

- 4+ years of overall experience with atleast 2+ years in AI products, voicebots, conversational AI or applied ML roles.

- Strong hands-on experience with prompt design, experimentation, and debugging.

- Deep understanding of voice AI systems (ASR, TTS, latency, turn-taking, interruptions etc.

- Ability to evaluate models and configurations pragmatically (not just theoretically).

- Comfort working with logs, system behavior, and ambiguous real-world issues.

- High ownership and bias for action.

You Will Have An Advantage If You :

- Have experimented with multiple LLMs and orchestration approaches.

- Have worked in delivery or implementation roles and understand real-world constraints.

Why should you be interested ?

- Influence both product direction and real-world system behavior.

- Get hands-on exposure to evolving LLM and voice AI ecosystems.

- Play a key role in shaping how AI systems perform at scale.