Founding AI Engineer
About the role
Founding AI Engineer at Upsolve AI, an AI-native BI platform startup. Role focuses on designing and shipping production-grade agentic LLM systems integrated with a modern BI stack, working closely with the CTO to drive product direction and deliver real user impact. Not a research role; high-impact production engineering with customer-facing outcomes.
Responsibilities
- Designing, prototyping, and deploying agentic LLM systems into production with real customer workloads.
- Running experiments to explore architectures, frameworks, and approaches and driving product direction from results.
- Building evaluation, observability, and improvement pipelines for agent quality and reliability.
- Integrating with the modern BI platform and customer-facing workflows.
- Establishing best practices for LLM/agent development, from tooling to performance monitoring.
- Embedding and collaborating directly with users and customers to validate production success.
Qualifications
- 7+ years of engineering experience, including deep experience shipping production LLM systems.
- 2+ years track record of building agentic or multi-step reasoning systems in complex, real-world environments.
- Strong command of modern AI/LLM tooling (LangChain, Mastra, DSPy, vector DBs, eval frameworks, observability, OpenAI/Anthropic APIs, fine-tuning).
- Strong engineering ability (TypeScript preferred).
- Experience with databases, data warehouses, data systems, data connections, and big data engineering/analytics.
Skills
Benefits
- Hybrid flexibility with central London office
- Competitive salary (£100k–£140k) plus founding-level equity (1–3%)
- Take-what-you-need holiday policy
- Ownership culture with low bureaucracy and high trust
About Upsolve AI
Embed GenBI in Your SaaS App
Upsolve AI is a customer-facing analytics as a service platform. We are building a full data stack that enables businesses to build and offer analytics to their customers at lightning speed and gives their customers the superpower to answer any data questions via AI. The company is founded by Ka Ling Wu and Serguei Balanovich, who built a similar product at Palantir before (featured in Palantir's S-1), growing it to 50+ enterprise customers and 8-figures of annual revenue in 2 years.