Senior Applied AI/ML Scientist (B2B)

NewsWhip by Sprout Social is part of the AI, Data, and Intelligence Business Unit. It processes the world's news in real time, turning millions of articles, posts, and signals daily into predictive intelligence used by journalists, PR leaders, and global brands. The role is focused on building and improving LLM-powered systems for news and media intelligence with high accuracy demands.

Minimum Qualifications:

• 6+ years building and operating production software systems.
• 2+ years hands-on experience shipping LLM-powered features.
• Strong backend engineering skills, preferably with Python.
• Experience with Large Language Model APIs, embedding models, vector databases, prompt engineering, LLM evaluation frameworks, and LLMOps practices using tools like Langchain, Datadog.
• Proven track record of owning end-to-end production systems.
• Experience making architectural trade-offs and leading technical reviews.
• Collaboration experience with Product and UX.

Preferred Qualifications:

• Bachelor’s degree in Data Science, Computer Science, Machine Learning, AI, or related field.
• Experience with AI orchestration frameworks such as LangChain, LlamaIndex.
• Familiarity with Model Context Protocol (MCP) or tool-calling architectures.
• Experience building agentic workflows and semantic search systems.
• Experience with cloud platforms and media analytics.
• Background in startup or high-growth environments.

Build and Ship LLM-Powered Features:

• Architect and deliver production agentic workflows for ambient and interactive agents.
• Own design of LLM-powered content enrichments feeding customer reporting and alerts.
• Lead technical direction for tool-augmented LLM systems including MCP-compatible services.
• Make architecture-level decisions regarding retrieval, fine-tuning, and model selection.
• Define structured prompting standards, output schemas, and reusable patterns.

Drive Quality, Evaluation, and Observability:

• Set standards for LLM evaluation including offline benchmarks and human-in-the-loop reviews.
• Own guardrails for safety, hallucination reduction, and output consistency.
• Build observability into LLM features with tracing, cost tracking, latency monitoring.
• Balance model quality, latency, and cost trade-offs.

Retrieval, Embeddings, and Semantic Infrastructure:

• Develop embedding and semantic search infrastructure, including chunking and hybrid search.
• Improve retrieval relevance as part of agentic architecture.

Lead Technically and Influence Cross-Functionally:

• Partner with Product and UX to deliver reliable AI features.
• Lead design discussions and represent AI team in architecture decisions.
• Provide mentorship and code reviews to engineers.
• Evaluate emerging models and techniques for inclusion in the stack.

Please be informed that the data controller is Spout Social. You have rights to your personal data including access, rectification, erasure, restriction, objection, portability, and complaint to supervisory authority. Personal data is processed for recruitment purposes based on legal obligations and consent. Data is processed until recruitment or consent withdrawal. Accommodation requests during recruitment can be submitted by email. Location flexibility is provided within Poland for remote or Kraków office work.