System Engineer

About In The Loop (ITL)We're In The Loop, a vertical software suite transforming the massive, overlooked world of secondhand retail. Thrift stores move billions of items every year, yet many still rely on handwritten tags, guess-based pricing, and fragmented systems. We're changing tht.Over the past year, we've partnered with some of the largest thrift operations in the country and have helped process 85,000+ garments per month. Now, with demand surging and new product modules shipping, we're officially in scaling mode.We're backed by leaders in the industry: including eBay, the former CTO of Depop, and the former CTO of Hepsiburada.What We BuildOur system plugs directly into the workflows of thrift processors, POS systems, and e-commerce marketplaces. We help stores go from:handwritten tags → automated listings,guess-based pricing → dynamic, data-driven pricing,lost revenue → recovered valueexternal, messy data dashboards → dynamic, daily viewsWhy This Space MattersMassive, broken industry: $211B re-commerce market and growing.Complex workflows: Every store and warehouse looks different, making it the perfect playground for problem-solvers.First movers: No one else is building a purpose-built production system for thrift.Clear ROI: Customers see velocity increases and revenue lift in months, not years.About the TeamWe're a small, product-obsessed team with deep experience in AI, resale, reverse logistics. We've built and scaled startup products from 0→1. At ITL, everyone wears multiple hats and is passionate about driving success for our customers. Systems Engineer The RoleWe are looking for someone who has shipped ML or computer vision systems in physical, operational environments and who is comfortable owning hardware and software together. If you have spent your career purely in web backends or SaaS, this is likely not the right role.What You'll Do / RequirementsComputer Vision & Physical Capture SystemsDesigning and deploying camera-based capture systems in real operational environmentsUnderstanding of optics, lighting, and image quality requirements for accurate CV inferenceExperience evaluating and specifying hardware — cameras, mounting, edge compute — not just consuming camera feeds in softwareFamiliarity with the gap between lab performance and production-floor reliabilityExposure to wearable or embedded capture systems is a plusML Infrastructure & LLM Pipeline EngineeringProduction-grade deployment of ML models and LLM pipelines on cloud infrastructureLLM orchestration: prompt design, caching, retry and fallback logic, provider routingLatency optimization across inference pipelines — identifying and eliminating bottlenecks from capture to prediction to UICost management at scale — token consumption, GPU instance economics, reserved capacity vs. on-demand tradeoffsTechnical Leadership & OwnershipLeading engineering team with accountability for outcomesCan set direction for other engineers clearly and without micromanagingMoves fast without leaving things broken — ships iteratively but with engineering judgmentCommunicates technical decisions to non-technical stakeholders clearlyOwnership orientation: you finish what you start and you care about whether it works in productionWhat We're Looking ForWe care about demonstrated evidence over credentials. The following are signals we weight heavily, regardless of years of experience.Non-negotiableYou have shipped a CV or ML system that runs in a real physical environment and is central to a client's operationsYou have made consequential architecture decisions and can speak to the tradeoffs clearlyYou are comfortable owning a problem end-to-end — from hardware spec to cloud deployment to production monitoringYou are based in Washington, DC or Northern Virginia and available to work in-officeStrong signalBackground in robotics, computer vision at the edge, warehouse or logistics automation, manufacturing systems, or any domain where ML meets physical environmentsYou have specified or evaluated camera hardware, capture rigs, or edge compute for a real deploymentYou have worked at a startup or small team where you had to make decisions without a playbook and live with the consequencesYou have optimized a high-throughput inference pipeline for latency or cost at production scaleExperience with frontier LLM APIs (Gemini, OpenAI) in production, including quota management and fallback handlingHow To ApplyEmail your resume tozahra@intheloopai.com