Senior Full Stack Software Engineer at Cynch AI in San Francisco, CA

Cynch AI is a Series A company that has grown revenue 4x in the past year. We build neuro‑symbolic AI applications for our accounting firm, Aardvark Tax Advisors, so our tax professionals can focus on the work and decisions that matter most. We are growing both organically and through strategic acquisitions, with a mix of remote and onsite employees and offices in San Francisco, the South Bay, and the East Bay.

We are looking for a senior software engineer who has fully embraced AI tooling and wants to stay at the forefront of how modern engineering teams move faster, build higher‑quality software, and solve more ambitious problems. This is a full stack role with a backend bias: you’ll work across the stack, but the hardest problems are in the data model, execution engines, and reliability—not just the UI or API surface.

The kind of work you may own here includes building execution engines for AI‑assisted tax workflows, document‑processing pipelines, data models for complex domain logic, systems for tracing and auditing automated decisions, integrations with tax/accounting platforms, and internal platforms that allow a small operations team to handle substantially more customer volume. You do not need prior tax or accounting experience, but you should have experience building production software where correctness, performance, and reliability really mattered.

We are changing how software can be built as we are building it; being all‑in on AI‑assisted development is central to this role.

What You’ll Do

• Work with founders, product, AI, operations and domain experts as you fully own substantial product and platform work from problem definition through production rollout.
• Build backend systems that can handle real production complexity: clear APIs, well‑modeled data, reliable execution, useful observability, and maintainable code.
• Turn messy operational workflows into clean software abstractions, internal tools, automations, and customer‑facing features.
• Continuously help the team discover where AI meaningfully improves software quality and velocity, and where it does not.
• Improve the engineering system itself: better tools, better patterns, better deployment practices, better observability, fewer repeated mistakes, and less accumulated technical debt.
• Mentor other engineers by raising the quality of design discussions, implementation choices, reviews, and production ownership.

How We Use AI

We are all‑in on AI‑assisted development, but with high standards for rigor. Your default process should integrate AI tools into your daily engineering workflows to continuously improve velocity and quality.

Engineers who thrive here use AI to explore designs, generate and test implementations, debug unfamiliar code, refactor safely, improve observability, and accelerate learning—while maintaining high standards for correctness, maintainability, and production quality. We are not looking for people who simply generate code and hope it works; we are looking for engineers who use AI to move faster because they already have the technical judgment to evaluate, constrain, test, and improve what it produces.

Technologies We Use

You do not need to know every technology we use, but you should be excited to work with a similar stack and learn quickly where needed.

• TypeScript, Python, Julia, Java, Go, Datalog
• Knowledge graphs, ontologies, neuro‑symbolic AI
• AWS EC2, ECS, RDS (postgres), Lambda, S3, Bedrock

Strong candidates often come from backgrounds such as data platforms, developer tools, workflow automation, compilers/languages, ML infrastructure, or enterprise SaaS platforms (fintech, tax/accounting software, healthtech) where backend systems must be correct, reliable, and scalable.

Who We’re Looking For

• Deep backend expertise and the ability to work across the stack, including frontend product experiences when needed.
• Has been directly responsible for a production system where correctness, performance, reliability, or scale created meaningful engineering complexity.
• Has designed data models, execution paths, background jobs, queues, retries, observability, and operational workflows for systems that had to keep working under real production load and real failure modes.
• Experience building systems where the core challenge was technical depth: workflow execution, rule evaluation, document processing, search/indexing, data pipelines, distributed jobs, domain‑specific language implementation, or correctness‑sensitive automation.
• Experience working in an early‑stage environment where requirements are incomplete, priorities shift, and ownership is broad.
• Strong product judgment: you ship code that actually solves the problem.
• You have operated at a level where you were trusted to own ambiguous, technically complex systems end‑to‑end, make architecture decisions, debug production issues, and raise the engineering bar for others.
• You can lead projects, take feedback, engage in thoughtful discussion, and get the work done.

We are looking for someone whose recent work goes beyond marketing sites, simple CRUD applications, prompt wrappers, prototypes, or frontend‑only product surfaces. The core of this role is building the underlying systems—data models, execution engines, observability, and controls—that make AI‑assisted workflows reliable, auditable, and useful in production, not just building UI around an LLM API.

How to Apply

As part of your application, please include a brief description (4–8 sentences) of the most technically complex production system you have built or owned. We are especially interested in the data model, algorithms, scale, reliability constraints, failure modes, and what you personally owned. Specific, concrete answers are much more useful than polished summaries; a little messy is much better than grand, generic language.

Medical, Dental, Vision, Life, LTD, 401k (with match)

Work Environment:

• Hybrid: San Francisco
• Open to Remote for Exceptional Candidates.
• California base salary range: $190,000–$240,000