Senior Software Engineer I — Agentic Analytics

Instacart

5d•CA$168,000 - CA$177,500•Remote

About The Position

We are currently seeking a Senior Software Engineer to join our Agentic Analytics Platform team — the team responsible for the AI-for-Data charter inside Instacart's Data Infrastructure org. You'll design and build LLM-powered systems that transform how data practitioners (data scientists, data engineers, analysts, PMs) interact with data at Instacart — from natural-language data access and AI-assisted SQL, to automated metadata generation, to embedding intelligent capabilities across our broader data infra ecosystem. This is a hands-on role at the frontier of applied AI inside a large, modern data stack. The mission of the Instacart Self-Serve organization is to improve the productivity of data practitioners through easy-to-use, self-serve tools. Agentic Analytics is the team chartered with bringing AI and LLMs into that mission — broadly. We own the systems, evals, and product surfaces that make AI-for-Data real at Instacart, and we partner across Data Infra to embed AI capabilities wherever they create the most leverage. It's a small team with an outsized charter, defining what "AI for the data org" looks like in production at company scale. You will be part of a small, ambitious team with a large amount of ownership and surface area, contributing directly to the AI-for-Data charter at Instacart. You will build LLM-powered systems end-to-end — retrieval/grounding over internal metadata and code, agentic workflows, evals and quality systems, and the product surfaces that put these tools in front of internal users. You will have the opportunity to suggest and drive new initiatives — the roadmap is still being shaped and there's real room for contribution. You will collaborate closely with other engineers on the team and contribute to a strong engineering culture around applied AI rigor (evals, observability, safety, cost). You will work closely with Data Science, ML, Product Engineering, and other Data Infra teams to identify high-leverage AI applications and translate them into reliable, production-grade systems.

Requirements

Self-motivation and an enjoyment for a startup environment, with comfort operating in ambiguity
A strong sense of ownership and a builder's instinct — you've shipped real systems, not just prototypes
Strong backend engineering skills in Python, with experience building production services
Hands-on experience with applied LLM systems — RAG, agentic workflows, prompt engineering, evals, or similar. You've worked on an LLM-powered system in production (or close to it) and have opinions on what makes them reliable.
Working knowledge of common data infrastructure technologies (Python, Airflow, Snowflake, dbt) in a production environment
Strong knowledge of SQL
Experience working with a large codebase on a cross-functional team
An ability to balance a sense of urgency with shipping high quality and pragmatic solutions
5+ years of working experience in a similar field/environment

Nice To Haves

Experience with retrieval systems, vector stores, or grounding LLMs over structured/unstructured enterprise data
Familiarity with eval frameworks and quality systems for LLM-powered products
Experience building developer tools or internal tools for technical users
Bachelor's degree in Computer Science, computer engineering, electrical engineering, OR equivalent work experience

Responsibilities

Build LLM-powered systems end-to-end — retrieval/grounding over internal metadata and code, agentic workflows, evals and quality systems, and the product surfaces that put these tools in front of internal users.
Suggest and drive new initiatives.
Collaborate closely with other engineers on the team and contribute to a strong engineering culture around applied AI rigor (evals, observability, safety, cost).
Work closely with Data Science, ML, Product Engineering, and other Data Infra teams to identify high-leverage AI applications and translate them into reliable, production-grade systems.