Senior Software Engineer

Job Type
Full-time
Description
The Opportunity
We're hiring a Senior Software Engineer to help build and scale our platform. This role is for engineers who work fluently alongside AI tools and agent frameworks — not as a novelty, but as a core part of how modern software gets built. You'll ship production systems, design architectures that scale, and raise the bar for code quality on a team where AI is a daily collaborator.
We don't want someone who just uses AI to type faster. We want someone who uses AI to think bigger — and who has the judgment to know when the AI is wrong, when it's right but unwise, and when to do the work themselves.
What You'll Do
Design, build, and operate production services that serve our customers — backend, infrastructure, and the integration points in between.
Lead technical projects from ambiguity to shipped feature: scope, design, build, deploy, measure, iterate.
Use AI coding tools (Claude Code, Copilot, Cursor, or whatever fits the task) as a force multiplier — to draft, refactor, explore, and review — while staying accountable for everything that ships under your name.
Build and integrate AI-powered features into our product where they create real customer value: agent workflows, LLM-backed APIs, retrieval systems, and evaluation pipelines.
Review code rigorously — including AI-generated code — and help establish standards for how the team incorporates AI output into the codebase safely.
Debug hard problems in production. Trace through systems you didn't build. Form hypotheses, verify them, and fix the cause, not the symptom.
Mentor other engineers, especially on the meta-skills that matter most now: judgment, taste, verification, and knowing when to push back on AI suggestions.
Contribute to architectural decisions and longer-term technical strategy.
Requirements
What we look for
Core Engineering Skills
7+ years of professional software engineering experience building and operating production systems.
Strong fundamentals: data structures, concurrency, distributed systems, databases, networking, security. The kind of depth AI tools can accelerate but not replace.
Production sense: you know what breaks at scale, how to add observability, how to roll out changes safely, and how to debug an incident.
Code review judgment: you can read a PR and spot the subtle bugs, the security holes, the architectural smells — including the ones AI tends to generate.
Clear technical writing: you can write a design doc that another engineer can build from, and you can explain a hard problem in plain language.
AI-native engineering practices
Fluent use of AI coding tools. You use them daily and you use them well — knowing when to delegate, when to verify, and when to write something yourself.
Healthy skepticism. You catch hallucinated APIs, subtly broken logic, and security mistakes in AI output. You don't paste-and-pray.
Experience with LLM-backed systems. You've worked with at least one of: agent frameworks (LangGraph, AutoGen, custom), tool/function calling, RAG pipelines, prompt evaluation, or model fine-tuning.
Understanding of failure modes. You understand the ways LLMs and agents fail in production: hallucination, drift, prompt injection, latency variance, cost blowups, and the operational practices that mitigate them.
Evaluation mindset. You measure model and agent quality with real metrics, not vibes. You've built or used eval harnesses, regression suites, or human-in-the-loop review processes.
Judgment and collaboration
Ownership - You take responsibility for the code you ship — whether you wrote it yourself or an agent drafted it. "The AI did it" is not an excuse you accept from yourself or others.
Taste - You can tell good design from bad. You push back on over-engineering and you push back on tech debt that compounds.
Communication - You translate fuzzy product requirements into concrete specs, ask the questions that surface hidden assumptions, and write things down.
Mentorship - You make the engineers around you better, especially at the new skills the field demands.
Nice To Have, Not Required
Experience building or operating production agent systems (multi-step, tool-using, with feedback loops).
Contributions to open-source AI tooling or agent frameworks.
Experience in EdTech domain.
Track record of mentoring or technical leadership beyond your own work.

Work arrangement
Yes