AI Developer

We're Avvoka Join us in building the next generation of legal drafting technology. We help teams move faster through complex contracts using automation, collaboration and AI — with lawyers firmly in control. Trusted by leading law firms, banks and enterprises, we've grown rapidly through product strength and word of mouth. Now we're scaling into a global platform — and rethinking how engineering itself gets done in an agentic world. Why Join Us This is a rare chance to help define what an AI-native engineering team looks like. You'll have the freedom to experiment, automate, and raise the bar for how software gets built in legal tech, alongside smart people who move fast and care deeply about quality. If you're motivated by ownership, pace and visible impact, you'll feel at home here. The Role We're looking for an AI Developer — an engineer who thinks in systems and builds with agents. You'll build in two directions at once: the product our customers rely on every day, and the internal tools and workflows that make our own team dramatically more efficient. You'll also help ensure the AI in our product is trustworthy - building the evals and guardrails that keep its output grounded and high-quality. Agentic engineering is new enough that nobody has a decade in it. But the gap between someone who's been building with these tools seriously for the past few years and someone just starting out is enormous. We're looking for the former: engineers with real depth in both AI and production systems, who've already formed strong, evidence-based opinions about what works and what doesn't. What You'll Do Build with agentic leverage — Design and ship features used daily by our users, orchestrating AI coding agents where they multiply your output and writing the hard parts by hand where they don't. You own the output either way. Multiply the team — Build the internal tools, agents and automations that make the whole engineering team faster. Design the harnesses, prompts and guardrails that let agents do real work across our codebase, not just toy tasks. Guarantee grounded AI output — Build and maintain eval suites that measure the accuracy, grounding and reliability of our AI features, and the guardrails that keep output faithful to source documents and safe for legal use. Catch quality drift before users do. Own performance and architecture — Architect systems that scale with our growing user base and keep them fast as they grow. What You'll Bring Strong analytical and systems thinking — you reason about data models, edge cases, failure modes and trade-offs, the way a good backend or systems engineer does. Real depth with agentic coding harnesses such as Claude Code, Codex or similar — you've been building with them seriously and have opinions grounded in real use. A rigorous, evidence-driven approach to quality — you can define what "good output" means and build the evals to measure it. Solid grasp of web application architecture, APIs, database design, and other aspects of modern development. Genuine enthusiasm for building the future of software with autonomous agents. Bonus points if You have production experience with Ruby on Rails (our current main app stack — a real plus, but not a requirement). You've worked with LLM evaluation, prompt engineering, or other AI grounding techniques. You've worked in SaaS, legal tech, or adjacent industries. 💡 If you’re excited about this role but your experience doesn’t align perfectly with every qualification, we encourage you to apply anyway — you might be just the candidate we’re looking for.