Sr. AI Engineer

Pendo•New York, NY

6h•Hybrid

About The Position

Emergent AI Products is a new team at Pendo, built from the ground up to explore, prototype, and ship AI-native experiences that change how software teams understand and serve their users. This is not an AI layer bolted onto existing product; it is a deliberate bet on what product intelligence looks like next. The team moves fast, operates with high autonomy, and builds products without clear precedents. As a Sr. AI Software Engineer, you will sit at the intersection of deep technical capability and strong product judgment. You will design and ship applied AI systems, including RAG pipelines, agentic workflows, and LLM-powered features, from prototype through production. You will make principled technical decisions, evaluate model behavior rigorously, and communicate tradeoffs clearly to engineers and non-engineers alike. This role is based in our New York office.

Requirements

Deep hands-on experience building and shipping LLM-powered systems, including retrieval-augmented generation, tool use, and agent orchestration frameworks.
Strong technical depth in system design, including choosing the right architecture, identifying failure modes early, and making tradeoffs that hold up across the product lifecycle.
Experience owning technical quality beyond your own features, including setting standards, catching problems in review, and improving shared infrastructure and tooling.
Strong command of model evaluation, including designing evaluation suites, reasoning about overfitting and bias-variance tradeoffs, and systematically detecting and mitigating hallucinations.
Solid understanding of modern model architectures, including transformers and diffusion models, with the judgment to decide when and how to apply them.
Production MLOps experience, including model deployment, monitoring pipelines, and latency, cost, and reliability optimization in a live environment.
Strong full-stack fundamentals with the ability to work across backend and frontend systems to ship complete, user-facing AI products.
Exceptional communication skills with the ability to explain complex technical decisions clearly to engineers, product managers, and executives.
Demonstrated product thinking, including the ability to ask whether something should be built before deciding how to build it.
You're a builder, not a maintainer.
You're most energized when there isn't a clear path yet, and you get to define it. You don't wait for direction; you identify gaps, shape solutions, and drive them forward.
You're AI-curious - genuinely.

Nice To Haves

Experience fine-tuning foundation models and a clear point of view on when fine-tuning outperforms prompting approaches.
Familiarity with AI safety considerations, guardrail frameworks, and responsible deployment practices.
Experience building agentic or multi-step reasoning systems using tools such as LangChain, LlamaIndex, or custom orchestration frameworks.
Background in a SaaS or product analytics environment where user behavior data informs AI design.
Prior experience contributing to or launching a net-new team or product area.

Responsibilities

Design and build AI systems including RAG pipelines, agentic workflows, and LLM-powered features, taking work from prototype through production and ensuring it can hold up in real customer environments.
Make principled decisions on when to prompt, when to fine-tune, and when to use a different tool entirely, explaining these tradeoffs clearly so the team can move quickly without sacrificing quality.
Instrument and evaluate model outputs rigorously by defining evaluation frameworks and catching hallucinations early, implementing guardrails that can withstand real-world load and production use.
Own model deployment, monitoring, latency optimization, cost management, and reliability at scale, helping to ensure AI systems are observable, efficient, and dependable in production.
Contribute across the stack when needed, working across backend and frontend to get AI-powered experiences in front of users.
Partner closely with product and design to frame problems well before writing code, pushing back when the framing is wrong and helping the team focus on what should be built, not just what can be built.
Stay current on the research and tooling landscape, including transformers, diffusion architectures, orchestration frameworks, and emerging agent patterns, bringing relevant advances back to the team and applying them thoughtfully.