Staff+ Software Engineer, AI - Aleph

About Aleph

Aleph is an AI-native platform for Financial Planning & Analysis (FP&A), tackling the challenge of scattered data, endless spreadsheets, and manual reporting in finance teams.

Founded by MIT and Google/Microsoft alumni and backed by top-tier VCs like Khosla Ventures and Bain Capital Ventures, Aleph works with customers such as Webflow, Notion, Zapier, and Y Combinator.

The Charter

Every AI feature at Aleph—from natural language financial queries to autonomous agents executing full FP&A workflows—runs on a shared foundation. This role owns that foundational layer end-to-end, including model selection and routing, context management, tool design, model proxy infrastructure, and evaluation systems.

What You'll Do

• Own the AI foundation product teams build on: model selection and routing, the model proxy layer, context management, and tool design.
• Own evaluation infrastructure in collaboration with finance domain experts to define and encode high-quality outputs.
• Ship agentic features end-to-end, from v0 through optimization loops including prompt/context engineering, caching, parallel tool calls, and subagent patterns.
• Build observability into agent behavior to profile performance, identify bottlenecks, and drive data-backed improvements.
• Track advancements in agentic systems and integrate proven practices into the engineering stack.

What We're Looking For

• Proven experience shipping production LLM systems (evals, context management, multi-model/multi-provider routing) with a strong grasp of past iterations and failures.
• Ability to rapidly build proof of concepts, make decisive architecture calls, and ship v0s driven by rigorous evals.
• Strong architectural judgment: opinionated about design and pragmatic about incremental shipping.
• Extensive experience with LLM APIs, agent frameworks, and user-facing products.
• A production-first mindset focused on shipping systems rather than pure research.

Staff+ Engineering Expectations

• Adoption: Systems you build are naturally adopted by teams you do not manage because of their high quality.
• Fundamentals: Ability to trace requests from UI to data store and make sound technical decisions at every layer.
• High Agency: Background as a founder, engineering lead, or startup builder with a creative mindset for shipping.
• Pragmatism: Choosing the fewest and simplest tools that will successfully do the job.
• Craft: Deep care for code quality, correctness, and maintainability.
• Communication: Clear, direct, and persuasive across both technical and non-technical stakeholders.
• Experience: Track record of building complex B2B products, ideally from scratch.