Founding LLM Engineer

We often joke that Fintool is "Warren Buffett as a Service." After Nicolas sold his previous startup—a legal search engine powered by AI—he invested part of the proceeds into Berkshire Hathaway stock, trusting in Buffett’s approach. Yet, with over a decade in AI, he couldn’t shake one question: Could a sophisticated language model replicate Warren Buffett’s investment process?

Warren Buffett's letters, biographies, and investment decisions provide a wealth of knowledge about how to find, analyze, and understand companies. There are even textbooks on value investing that detail the step-by-step process.

What if we could break down Buffett’s process into individual tasks and use an AI agent to replicate his approach?

At Fintool, we took on that challenge. We deconstructed most of the tasks that Buffett performs to analyze a business—reading SEC filings, understanding earnings, evaluating management decisions—and we built an AI financial analyst to handle these tasks with precision and scale.

Nicolas Bustamante: spent 7 years building one of the largest AI-driven legal search engines (Bloomberg for lawyers). Nicolas hired nearly 200 people, secured millions of dollars in debt and equity funding, and the profitable business was successfully acquired by Summit Partners, a $43B billion growth equity fund, for $x00M+

Edouard Godfrey: worked for 9 years at Apple, leading teams of data scientists and engineers. He worked on Apple Search (Spotlight) and Apple Pay, maintaining big data pipelines and deploying cutting-edge AI models. He received the 2019 Apple Pay Innovation Award for outstanding contributions and fresh insights.

Click here to see our full LLM tech stack.

Small team: small in-person teams outperform large and well-funded companies. When people visit our office, they should be surprised by how few people we are.

Ship code: we avoid meetings, PM jargon to release early, release often, and listen to customers.

In-person: we believe high-performing teams do their best work, build long-term relationships, and have the most fun in person.

Clone and improve the best: we're not about reinventing the wheel but about enhancing proven success. We are shameless cloners who stand on the shoulders of giants. We draw inspiration and then create differentiation because distinctiveness drives dominance.

Release early, release often, and listen to your customers: speed matters in business, so we push better-than-perfect updates for customers asap. Mastery comes from repeated experiments and learning from mistakes rather than putting in a set number of hours. It’s 10,000 iterations, not 10,000 hours.

Warren Buffett: We model our personal and professional ethos on the principles he exemplifies. Upholding integrity, valuing honesty, practicing frugality, championing lifelong learning, embracing humility, extending generosity, applying rationality, and demonstrating patience. Every day, we strive to mirror these Buffett-inspired virtues.

Your challenge is to develop a high-quality, low-latency retrieval augmented generation (RAG) on top of millions of complex documents. You will lead the implementation of custom embeddings, rankers, and hybrid search.

Requirements: Python, Machine Learning (embeddings, ranking, recommendations), LLM, . Knowing Spark, Databricks and Typescript is a plus.

Experience: 3+ years

Location: San Francisco (no remote)

Contract: Full-time

Apply via the form below 👇