The company builds a tax automation platform that lets software businesses sell worldwide without the compliance burden.
Digital commerce has grown 6x over the past decade, and software firms now have to track sales tax exposure, work out rates and file returns across more than 20 US states and many other countries. That gap blocks companies from selling to customers anywhere.
The platform connects to billing and payment systems to automate tax monitoring, calculation and filing. It covers municipal, state and federal rules, so customers can focus on growth.
40% of the Forbes Top 50 AI companies and 20% of the Forbes Top 100 Cloud companies are customers, alongside firms such as Notion, Anthropic and Cursor.
Responsibilities
The team is not growing by adding analysts and reviewers for every task. It is automating the routine work.
As the first security engineer, you will build the product and cloud security program across secure development, a hardened cloud, detection and response, and incident leadership. Agents take the repetitive tasks so you can make the judgment calls. Your first-year work will decide how far a small team can cover as the company grows.
The company sits in customers' revenue flow and holds sensitive financial, tax and identity data, so security is part of what customers buy.
Repeated tasks become automated workflows. Agents propose, and a person approves anything that changes production, closes a finding or reaches a customer, with an audit trail. Each agent has its own identity, least-privilege access, logged actions and evals before use. Success is measured by how much coverage one engineer provides, not by tickets closed.
Product and application security: add scanning to CI, run AI-assisted review on pull requests that touch payments, credentials or auth, and threat-model the highest-risk designs. Triage most findings automatically, run the pentest and vulnerability disclosure programs, and agree each fix with the owning team.
AI security: secure the AI features, including MCP and agent integrations, covering tool permissions, prompt injection, data boundaries between customers and testing before each launch. Set guardrails for internal AI use, covering which tools reach which data, coding agents, AI vendors and internal tools built by non-engineers.
Cloud and infrastructure: maintain a hardened baseline written as code and checked continuously, with secrets that rotate automatically. Keep access to customer data tight across production and internal tools, with an owner and scoped permissions for every service account, API key and agent identity.
Detection and response: build detections across cloud, identity and source control, with an AI triage layer that investigates routine alerts and escalates the important ones. Run the incident response plan and lead incidents from triage through postmortem.
Across the team: build SOC 2 and ISO 27001 technical controls with the GRC Lead so they generate their own evidence. Support enterprise security reviews and help shape the Security team's roadmap and hiring.
Within 1 year: every critical and high finding has an owner and is fixed within SLA, even as engineering doubles. Agents handle most routine alerts and findings, and the share is measured. Every agent has an owner, scoped access, logs and evals. Employees use AI tools within guardrails that keep customer data protected, and every AI feature is threat-modeled and tested. No standing human admin access to production, and secrets rotate automatically. The company passes its pentest and its SOC 2 and ISO 27001 audits, with controls that produce their own evidence. Engineers bring you in early because working with you makes them faster.
Requirements
6+ years in security engineering, with hands-on work in application security, cloud security, and detection and response.
You have built with AI for real security work, such as agents, LLM workflows, MCP servers or retrieval over your own data. You understand how these systems fail, including made-up findings, prompt injection and overly broad permissions, and you design against those failures. Share an example of something you have shipped.
You have locked down access to sensitive data across production and internal tools, including scoped identities for people, services and agents.
You have owned technical controls for SOC 2 or ISO 27001 and prefer evidence generated by automation over screenshots.
You code well, read production code fluently and can review a pull request for security rather than only running a scanner.
You have hands-on security experience in a major public cloud and are fluent in infrastructure as code and CI/CD, such as Terraform or GitHub Actions.
You have led security incidents and can run a calm, well-documented response.
You judge risk well, focus on the controls that matter most, explain trade-offs clearly to engineers and executives, and make the secure path the easy one for engineering teams.
You have been the first or an early security engineer at a fast-growing startup.
You have secured LLM features, AI agents or MCP integrations in production, including evals or red-team testing.
You have rolled out company-wide AI guardrails, such as tool and connector policies, shadow-AI discovery or data loss prevention.
You have owned SOC 2 or ISO 27001 technical controls with evidence from automation rather than screenshots.
You have done detection engineering as code.
You have worked in fintech, payments or B2B SaaS handling sensitive financial data.
You have run pentest, vulnerability disclosure or bug bounty programs.
Certifications are welcome but not required. The company cares more about what you have built.
Conditions
Team offsites and gatherings take place at the San Francisco headquarters office throughout the year.
Benefits
Equity upside of an early-stage startup, with the product-market fit of a later-stage company.
Medical, dental and vision insurance, covered at 100%.
One Medical membership covered, flexible sick benefits and more.
Annual learning and development stipend for books, online courses and conferences, plus a curious team to learn from.