Software Developer – Backend (AI-Powered CAD Startup)

Advatix, Inc.•San Mateo, CA

1d•$110,000 - $170,000•Onsite

About The Position

This backend engineering role requires a rare blend of production-grade software development and deep academic research expertise. You will own the core platform end-to-end, designing and maintaining scalable backend services in Python while developing the APIs that power sophisticated AI-driven CAD workflows. Your daily work involves building robust systems using Flask or Django, optimizing SQL schemas, and deploying production services within cloud infrastructure. Beyond standard backend tasks, you will be responsible for implementing and integrating custom machine-learning models into production features, requiring you to translate complex research into reliable, high-performance software. The ideal candidate possesses significant research depth, typically evidenced by a PhD from a top-tier program or a substantial body of novel systems work including at least seven peer-reviewed publications at venues such as NeurIPS, ICCV, or ICML. You must be capable of writing custom PyTorch implementations of novel modules rather than relying on existing wrappers. This position is designed for a product-minded builder who thrives in a flat, high-caliber founding team and prefers hands-on technical ownership over managerial titles. You will participate in the full development lifecycle, from early technical design through launch, ensuring that every feature is fast, reliable, and capable of handling the rigorous demands of hardware engineering.

Requirements

Strong Python and modern software engineering practices (testing, code review, version control)
Experience building backend services and APIs (Flask, Django, or similar)
Working knowledge of SQL and relational databases — able to design, query, and optimize schemas
Comfortable with cloud infrastructure (AWS, GCP, or similar) and deploying production services
Track record of taking research or prototype code into reliable, production-grade systems
PhD from a top ML or systems program (Michigan, CMU, Stanford, MIT, Berkeley, Georgia Tech, UIUC, or equivalent), or an equivalent body of novel systems work
Published research at top venues (NeurIPS, ICCV, ICML, IEEE) — 7+ peer-reviewed publications is the benchmark
Custom PyTorch implementations of novel modules — not wrappers around existing libraries
Hands-on edge deployment experience
US citizen or existing work authorization that does not require future sponsorship
Based in the US with existing work authorization; US citizen/visa only, and able to work primarily on-site in San Mateo, CA.
Hands-on experience with SQL and relational databases in a production or serious project setting.
Strong proficiency in Python and experience building backend services with Flask or Django.
Published peer-reviewed research (NeurIPS, ICCV, ICML, IEEE, or equivalent) or demonstrated body of novel ML systems work — LLM/API integration does not qualify.

Nice To Haves

Product-minded builder who loves turning ideas into reliable, maintainable code.
Fluent in Python, comfortable with backend frameworks like Flask or Django, and you care about performance, correctness, and clean abstractions.
Curious about how ML can power better tools, even if you’re not an ML expert yet, and you’re excited to work closely with founders and customers to ship the right thing — not just more lines of code.
Likes small teams, high ownership, and the idea of spending the next few years working on a single, hard problem: making AI-powered CAD the new standard for hardware design.

Responsibilities

Design, build, and maintain robust, scalable backend services in Python.
Develop APIs and backend components that power Hestus' AI-driven CAD workflows and improve overall system performance.
Collaborate with founders and cross-functional partners to define, design, and ship new product features.
Participate in code reviews, testing, and debugging to ensure high-quality, reliable software.
Implement and integrate machine-learning models and algorithms into production features (with support from the ML team).
Diagnose and resolve production issues; improve observability and reliability over time.
Contribute across the full development lifecycle, from early technical design through launch and iteration.
Stay current on relevant backend, cloud, and ML tooling and bring practical, modern solutions back to the team.