> Markdown version of [/jobs/ext/2716874-agent-data-scientist](https://www.wearedevelopers.com/jobs/ext/2716874-agent-data-scientist). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Agent Data Scientist - **Company:** DECAGON, LLC - **Location:** San Francisco, CA, United States - **Experience:** Expert - **Salary:** $41,600.0 - $166,400.0 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Python (Programming Language), SQL Databases, Large Language Models - **Published:** September 4, 2026 - **Apply:** https://startup.jobs/agent-data-scientist-decagon-9615793 ## About the Role * 5+ years in data science, analytics, or another quantitative role where you drove analyses end to end and used them to move decisions, whether at a high-growth technology company, in consulting, or in a forward-deployed or solutions role. * Exceptional judgment when interpreting metrics. You think through the edge cases quickly, and you can see how the same number will be read by someone arguing for us and against us. This includes 80/20 intuition on what's most important vs. analyzing everything. * Comfort in fast-moving, ambiguous situations where you shape the approach as much as you implement it. Given a question with no established right answer, you can identify the best available option and defend it. * Strong independence. You operate with very little day-to-day direction, set your own priorities against a broad mandate, and bring decisions rather than questions. You are comfortable being the person who defines the work rather than the person who receives it. * Ability to bring stakeholders to a shared definition of success, even when they arrive with different priorities and different levels of technical depth. * Strong technical fundamentals. Fluency in SQL and Python, and comfort working with large volumes of messy, unstructured conversational data. Even better if you have * A history of building internal tooling that gained meaningful adoption, and the judgment to recognize when a problem should stay a one-off. * Experience working on AI products in production, including evaluating LLM or agent behavior and building the evaluations yourself. ## Description Most of your time will be spent working directly with customers to establish what should be measured and then building what produces it. Usually, the hard part comes from driving alignment, not the analysis itself. You own both reporting on performance and improving it, working alongside the Agent Product Managers, Agent Strategy Managers, Strategic Account Directors, and Agent Deployment Engineers on each account. Beyond the work you do with individual customers, you will build the product features and frameworks that make everyone else at Decagon better at this work. This is a rare combination: the rigor of a strong quant, the presence of a trusted executive advisor, and the ownership of a founder. If you have ever wanted to own the metric rather than report on it, this role was designed for you. This is a senior individual contributor role. You are trusted to make high-stakes decisions independently and own outcomes. In this role, you will * Work directly with customers to define the metrics that will be used to measure performance, then operationalize them so that both sides use and trust them. * Own the delivery of those metrics, where transparent access to data helps reinforce the narrative we build with the customer. * Own improving the metrics that matter most on your accounts, from diagnosing what drives performance to designing the experiments and driving the changes that move them. * Scope and build the product features, tools, and frameworks that empower the rest of Decagon to do more of this work themselves. * Define what success looks like for new initiatives that have no precedent. As Decagon's product continues to evolve, new surface areas will emerge where the right way to measure success has to be invented rather than applied. * Run tight feedback loops into Product, Engineering, and Research, shaping the product based on what you learn measuring our most demanding deployments. ## Related Videos - [Fault Tolerance and Consistency at Scale: Harnessing the Power of Distributed SQL Databases](https://www.wearedevelopers.com/videos/1146-fault-tolerance-and-consistency-at-scale-harnessing-the-power-of-distributed-sql-databases) - [Creating Industry ready solutions with LLM Models](https://www.wearedevelopers.com/videos/899-creating-industry-ready-solutions-with-llm-models) - [Edit Your Future: Queerverse Radical AI](https://www.wearedevelopers.com/videos/909-edit-your-future-queerverse-radical-ai) - [Why Your AI Agent Keeps Hallucinating Your Data: Building Deterministic Context Layers](https://www.wearedevelopers.com/videos/2055-why-your-ai-agent-keeps-hallucinating-your-data-building-deterministic-context-layers) - [Fault Tolerance and Consistency at Scale: Harnessing the Power of Distributed SQL Databases](https://www.wearedevelopers.com/videos/1520-fault-tolerance-and-consistency-at-scale-harnessing-the-power-of-distributed-sql-databases) - [Data Science in Retail](https://www.wearedevelopers.com/videos/586-data-science-in-retail) ## Related Articles - [Graph and AI Trends 2026: Why Is AI Running but Not Yet Delivering?](https://www.wearedevelopers.com/magazine/680-graph-and-ai-trends-2026-why-is-ai-running-but-not-yet-delivering) - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [Coffee with Developers - Maria Apazoglou - Making AI understandable for all in production](https://www.wearedevelopers.com/magazine/475-coffee-with-developers-maria-apazoglou-making-ai-understandable-for-all-in-production) - [Navigating the AI Shift](https://www.wearedevelopers.com/magazine/629-navigating-the-ai-shift) - [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care) - [MLOps And AI Driven Development](https://www.wearedevelopers.com/magazine/82-mlops-and-ai-driven-development)