AI Product Analyst Intern

About the Role

We are looking for an AI Product Analyst Intern who enjoys digging into data, understanding AI systems, and solving product problems. You will work closely with the Product, AI Engineering, and Business teams to analyze how our AI agents perform in production, identify where conversations or workflows break down, measure product health, and build dashboards that help teams make better decisions. This role is intentionally broader than a standard product analyst: you become the health monitor for our AI platform, tracking everything from model quality (STT, TTS, SLMs, LLMs) and telephony reliability to product workflows and business KPIs, then turning those insights into actionable improvements for engineering and product teams. It is an ideal role for someone interested in Product Management, AI, Analytics, and Data.

Key Responsibilities

• Analyze production AI conversations across Voice AI and In-App AI Agents.
• Identify where users drop off or where AI responses fail.
• Investigate failures across the entire AI pipeline: Speech-to-Text, Text-to-Speech, voice-to-voice models, Small Language Models, LLM reasoning, retrieval and memory, the telephony pipeline, and Action Intelligence.
• Work with Product and AI teams to identify root causes of quality issues.
• Track product health metrics and recommend improvements.
• Build dashboards for Product, Business, and Engineering teams.
• Define and monitor KPIs across different AI systems.
• Analyze customer journeys and identify friction points.
• Measure conversation quality, latency, task completion, and automation success.
• Generate insights from production data to improve customer experience.
• Assist in A/B testing new AI models and product features.
• Document recurring issues and suggest long-term improvements.

Example Problems You'll Solve

• Why are users abandoning a conversation after 30 seconds
• Which STT model performs best across different accents
• Which TTS voices produce the highest engagement
• Why did an AI workflow fail to complete
• Which telephony providers have higher call failure rates
• Where are AI agents hallucinating or misunderstanding user intent
• Which product flows generate the highest drop-offs
• Which enterprise customers experience the most AI failures
• Which AI models should be replaced or fine-tuned
• What are the biggest contributors to latency across the AI pipeline

Dashboards You'll Build

You'll design and own dashboards including:

• AI Quality and Product Health
• Voice Pipeline and Conversation Analytics
• Customer Journey and Model Performance
• Telephony Reliability and Business KPIs
• Error and Failure Analytics
• AI Cost vs Performance

Requirements & Tech Stack

• Strong analytical and problem-solving skills.
• Comfortable working with large datasets.
• Basic SQL knowledge.
• Python for data analysis (preferred).
• Understanding of product analytics.
• Familiarity with dashboards (Metabase, Power BI, Tableau, Looker Studio, and similar).
• Curiosity about AI and large language models.
• Strong communication skills and the ability to work with cross-functional teams.

Bonus Points

• Experience with AI or LLMs.
• Understanding of speech technologies (ASR, STT, TTS).
• Knowledge of prompt engineering.
• Experience with conversational AI.
• Familiarity with product analytics tools such as Mixpanel, Amplitude, or PostHog.
• Basic understanding of APIs and backend systems.

Qualifications

Students or recent graduates pursuing Computer Science, Data Science, AI/ML, Analytics, Information Systems, or related fields who enjoy understanding how products work, analyzing data, and improving AI systems.