The posting, in Sprout Social's own words
archived Sep 29, 2026Description
Sprout Social is looking for a Senior Applied AI/ML Scientist to join its AI, Data, and Intelligence Business Unit and drive advancements to our suite of agentic and ML capabilities that are the foundation of Sprout Intelligence. Sprout Social empowers businesses worldwide to harness the power of social media. Processing over one billion social messages daily, our platform serves essential insights to over 30,000 brands. We are weaving AI throughout our products, and this role sits at the center of making that AI trustworthy: as we ship agents across our product suite, the hardest and most valuable problem is knowing whether an agent is actually good, and making it better. That is the work of this role.
The Role
Own the development of agents and agentic features alongside our engineering teams. Scientists define what quality means for our agents, design and calibrate the evaluations and judges that measure it, decide which models to adopt and why, identify and implement improvements to core agentic capabilities, and hold the quality bar as we scale Sprout’s agentic platform. If you are energized by the question "how do we effectively use AI to solve hard problems for our customers”, then this is the role for you.
Read the full posting ↓
What You'll Do
Own end-to-end agent and agentic feature development, including defining requirements, definitions of success, evals, guardrails, the necessary tools; and working with engineers, designers, and PMs to execute. Lead model selection, development and validation: rigorously assess whether a model or configuration is genuinely better, and make well-reasoned adoption decisions. Design and run rigorous experimentation (A/B testing and other methods) to measure the real impact of AI features on customers. Set and evolve the quality bar across agents, and partner with engineers who build and run the agents and the eval infrastructure so your judgment scales across many systems rather than one. Identify opportunities to improve the capabilities of our agent platform, ranging from prompt iteration, reusable judges, context window management, tool use, adopting new state of the art models. Communicate AI capabilities, limitations, and quality clearly to product, design, engineering, and leadership, so decisions are grounded in evidence.