Job description
Kanary | Full Stack Engineer | Remote (US) | Competitive Salary + Equity + Benefits | Python / React Kanary is a personal security platform that monitors the web to identify and reduce risk based on an individual's threat model. We've been building since 2020, started in campaign security and data science, backed by Mozilla and Y Combinator. We have 200+ organizations on the platform and are funded by our customers. Attackers have never had better tools. LLMs, social media, and search surface personal data in new ways. A few prompts can launch a deepfake and impersonation campaign. Risk sources appear and disappear across social platforms, telegram chats, and private data broker platforms. Kanary exists to play the cat and mouse game for people at the speed and scale of an agentic internet. Kanary is a combination of software workflows (crawling, scraping, automation), AI (data classification, research, code generation), and people (analysts trained to support customers and escalate). This is an exciting opportunity for those who are interested in building systems and agents to solve hard and often adversarial problems. We’re growing our team, and seeking a multifaceted engineer (2+ years experience). We’re looking for engineers with web scraping, data pipelines, or security tooling experience. Experience with Python/Django, Typescript, React, and LLM APIs is a plus. Must be authorized to work in the US. Email steven@kanary.com with something you've built and why it was hard.
✍️ Tailored application (cached — grounded in your resume)
72 / 100 honest fit
Daniel is a strong backend/Python match with unusually direct experience in the two things this role leans on hardest: shipping full-stack Python web services end-to-end, and doing it with LLM/agentic tooling as a core part of the workflow. Independently he has designed, deployed, and operates production Flask apps on a self-hosted Linux server with hardened endpoints, per-IP rate limiting, automated cron data pipelines, and self-healing watchdogs serving real organic traffic — closely adjacent to Kanary's crawling/pipeline/automation stack — and his day job in regulated medical-device software makes reliability a requirement rather than an aspiration. The obvious gap is frontend: his resume shows no React or TypeScript, and his web work is Flask/Python rather than Django + React, so he'd be ramping on the frontend half of "full stack" as defined here.
Tailored resume highlights
- Designed, built, and operate a suite of production Flask/Python web applications on a self-hosted Linux server — live public products behind Cloudflare tunnels (gunicorn, hardened endpoints, per-IP rate limiting, automated cron data pipelines, self-healing watchdogs, SEO) serving real organic traffic; idea to deployed, monitored production service in days.
- Use LLM / agentic tooling (Claude Code) as a core part of the engineering workflow, not a demo — shipping production full-stack services end-to-end, including a multi-strategy algorithmic trading platform with brokerage API integration, automated backtesting, a live dashboard, and reboot-safe scheduled execution.
- Build and optimize backend services and data/validation pipelines for connected medical devices, with AWS-based processing and monitoring (Lambda, S3, DynamoDB) for scalable, reliable data delivery.
- Lead root cause analysis of complex field-found issues — reproducing rare, intermittent defects in-house and driving resolution across multiple teams; the same adversarial debugging instinct a cat-and-mouse problem space rewards.
- Implemented secure C and Python APIs with authentication and data access control for IoT endpoints, and develop Python automation frameworks wired into Jenkins CI/CD for reproducible deployments and regression coverage (cut manual test effort 40%).
"Why this company?" (draft — personalize before sending)
Kanary's combination — scraping and automation workflows, AI classification, and analysts in the loop — is close to what I already build on my own time. Over the past year I've shipped and operated roughly a dozen production Flask services on a self-hosted Linux server using Claude Code as a core part of the workflow: cron data pipelines, hardened endpoints, per-IP rate limiting, self-healing watchdogs, real traffic. That's agentic development as daily practice, not a demo. My day job is connected medical devices, where being wrong has consequences — and personal security is the same kind of problem. I'd like to point both at an adversarial one.
Cover letter (review before sending)
Dear Kanary team,
I'm applying for the Full Stack Engineer role. Your description of Kanary as software workflows, AI, and trained analysts working together describes the kind of system I already build.
At Medtronic I develop backend services and validation pipelines for connected medical devices, Python automation frameworks wired into CI/CD, and AWS-based processing on Lambda, S3, and DynamoDB. A lot of that work is root-cause analysis — reproducing rare field-found defects in-house and driving fixes across teams. Regulated medical-device software makes reliability a requirement, not a feature.
Independently, since 2025 I've designed, shipped, and operated a suite of production Flask/Python applications on a self-hosted Linux server: public products behind Cloudflare tunnels running gunicorn, with hardened endpoints, per-IP rate limiting, automated cron data pipelines, and self-healing watchdogs, carrying real organic traffic. I built them rapidly using LLM/agentic tooling (Claude Code) as a core part of my workflow, including a multi-strategy algorithmic trading platform with brokerage API integration, automated backtesting, and a live dashboard.
One honest gap: my production frontend work is Flask-based, not React/TypeScript. I'd ramp there fast.
Daniel Maynard
ATS keywords you have but should add
Full StackFlaskData pipelinesLLM / agentic tooling (Claude Code)Linux (self-hosted server administration)gunicorn / CloudflareAuthentication & access controlRate limiting / hardened endpointsCron / scheduled jobsProduction monitoring
Everything is grounded strictly in your resume — review before you submit. Nothing is auto-sent.