Software Engineer (New Grad)

About Saris AI

We're a San Francisco, Toronto, & Montreal based applied AI startup that's building the future of work in the banking industry. We are tackling a $100 billion/yr problem, doubling every quarter and pushing the boundaries of what's possible with multi-turn AI agentic systems.

Our goal is to tackle the type of automation problems that require long-context reasoning, tool orchestration across legacy systems, and strict compliance loops: the ones without known answers.

We've shipped real agents that handle real customer workflows in production. With a growing customer base and live deployments, we're scaling up fast and looking for deeply technical builders who want to have outsized impact early.

Our core engineering team is hiring new-grad Software Engineers who are already fluent with agentic coding tools and want to put that fluency to work on real production systems from day one. You've spent your undergrad shipping things — class projects you took past "demo day," internships where you owned features, hackathons where you built something that worked, side projects with actual users — and you've been living in Claude Code, Cursor, or similar tools long enough to have strong opinions about how they should be configured. This role is in-person in our San Francisco, Toronto, or Montreal office.

Your mission is to

• Build, test, and ship AI-powered features that handle real banking workflows in production.
• Own features, components, agent loops end-to-end
• Use agentic coding tools as your default mode of working, and help us define what "good" looks like as we scale around AI-native engineering practices.
• Learn the regulated-finance domain deeply. Our customers care about getting things right, and so do we.
• Contribute to code reviews, testing, and engineering practices; raise the bar where you can see how.

Who You Are

• Completed an undergraduate degree in Computer Science, Computer Engineering, Electrical Engineering, Math, Physics, Statistics, or Applied Math
• Real builder track record: internship work you owned, class projects you took seriously, side projects with users, hackathon wins, or OSS contributions we can read.
• Strong fundamentals in Python (Django, Flask, or FastAPI) and/or TypeScript/React, we want to see confidence in at least one stack.
• Comfortable with PostgreSQL, Docker, and at least one cloud platform (AWS or GCP). Production-scale experience not required; working understanding is.
• Deep practical experience with agentic coding tools (Claude Code, Cursor, Aider, etc.)
• Excited to work in-person in our SF or Montreal office.

Bonus Points If You

• Have shipped a side project that real users actually use (not a TODO app you abandoned in week 2).
• Have contributed to an open-source agent framework, eval tool, or LLM-adjacent library.
• Have built something with multi-step LLM workflows, tool-calling, or eval loops that worked end-to-end.
• Have written publicly about how you actually use agentic tools day-to-day.

Why Join Saris AI?

• 🏦 Join us in building the future of work for the trillion-dollar banking industry using cutting edge AI technology.
• ⚡ Tackle ambiguous technical challenges with no clear answers.
• 🚀 Get production ownership in year one that takes 3–5 years to earn at a bigger company.
• 🧠 Work alongside senior engineers who actually have time to mentor you — not buried under sprint planning for a 200-person org.
• 💲 Competitive compensation with premium benefits and equity package.
• 🤝 Work with a stellar team of engineers, builders, and leaders; including repeat YC founders with a successful exit (Ready Education).
• 📈 We already have production agents live with revenue-generating customers.
• 🔥 Our team is backed by Tier 1 Silicon Valley VCs.

Rules File

As part of your application, submit the rules/config file you actually use day-to-day with your agentic coding tool of choice (ex: CLAUDE.md, AGENTS.md, .cursorrules), your system prompt, your custom slash commands, whatever shape it takes. We want your real one, not a polished version made for this application.

Plus two paragraphs max on the rules or sections you consider most valuable and what made you add them.