AI Product Engineer II, Assistant; Full Stack

Amplitude is the leading digital analytics platform that helps companies unlock the power of their products. Over 4,300 customers, including Atlassian, NBCUniversal, Under Armour, Square, and Jersey Mike’s, rely on Amplitude to gain self-service visibility into the entire customer journey. Amplitude guides companies every step of the way as they capture data they can trust, uncover clear insights about customer behavior, and take faster action.

When teams understand how people are using their products, they can deliver better product experiences that drive growth. Amplitude is the best-in-class analytics solution for product, data, and marketing teams, ranked #1 in multiple categories in G2’s Summer 2025 Report. Learn how to optimize your digital products and business  .

As an organization, we approach challenges with humility, take ownership of our contributions, and embrace a growth mindset that pushes us to constantly improve ourselves, each other, and the value we bring to customers and partners.

Amplitude’s Commitment to Diversity Equity & Inclusion (DEI):Amplitude believes that diversity enables the creation of better products, improves the ability to solve complex problems, and drives more powerful solutions. We strive to create an environment of inclusion—one focused on psychological safety, empathy, and human connection—that will allow employees of all backgrounds to thrive.

What we’re working on:

We’re a team of startup engineers working on a new product within Amplitude.

Here’s what drives us:
It’s crazy how little software has changed over the past 20 years. It’s gotten a lot more powerful, but users still mostly do the same things – find the right page, find the right button, google how to do something, etc. We believe software should flex to meet users where they are rather than the other way around.

We’re building an AI-powered personal chaperone for users, embedded into websites and apps. It’s a chatbot, but it’s a lot better in a few ways:

• It uses AI to personalize responses to every user, leveraging the power of Amplitude data
• It can show users around with a cursor – think “tour guide” mode
• It can perform actions on behalf of users by using APIs and interacting with other agent

We’re looking for an excellent product builder to join us - someone who’s as excited as we are to build the next generation of UX and bring it to millions of users.

What makes this role special?

• Immediate impact: The product you’ll be working on has real customers that are getting real value from it - no AI vaporware.
• High velocity: No sitting in meetings waiting for direction - we make decisions quickly and are always shipping. Our team came over from Command
  
  AI and we embody startup energy.
• Killer team: We have >3 years of experience building production AI products. You’ll get to learn how real-world AI applications are made, deployed, monitored and scaled.
• Data moat: Amplitude has the largest repository of behavioral data in the world.
• We keep it light: We build cool stuff and have fun while we’re doing it. You’re someone who enjoys working with other curious people.

What will you be doing?

• Owning the product: Our team is hyper-focused on building a great product - this includes using the product, listening to customer feedback, and surfacing ideas for where we can improve. There’s no engineering vs. product fences here - everyone is a product owner.
• Shipping features: Own full-stack features end-to-end: from ideation to design to implementation.
• Exploring the frontier: AI is constantly evolving - you’ll be testing out new models, techniques and libraries to chart the path forward.

Examples of everyday work:

• Pair with our AI Engineering Lead to debug why an agent gave out a bad answer — and design a clever retrieval fix that solves the edge case.
• Build a pipeline that automatically spots blind spots in a knowledge base and automatically fills in the missing piece.
• Experiment with new frontier models and libraries, then present back to the team on what’s hype vs. what we should ship next.
• Jump into a customer debug session with a PM, see how an agent failed in the wild, and…