KorrAI

Why we're building TRAIL We started KorrAI to monitor ground deformation on critical infrastructure using satellite data. Millimeter-scale precision on where the earth moves. But working with operators and engineers, we hit the real constraint: synthesis. A closure engineer inherits decades of scattered data. Satellite imagery, water monitoring, geology reports, soil surveys, historical records, regulatory filings. They spend weeks manually reconstructing site history before they can assess remediation options. Regulators re-ask the same questions because there's no traceable record of what was checked and why. The bottleneck isn't data availability. It's operationalizing all of it: building a workspace where closure teams can ingest multi-modal sources, surface what matters, and produce reports that cite everything. We call it TRAIL. The thesis Billion-dollar infrastructure decisions rest on scattered evidence. Mining closure, construction risk, data center siting are high-stakes, high-evidence environments. Being wrong costs regulatory delays, remediation overruns, or worse. AI won't replace closure engineers or regulators. But AI can do what humans shouldn't: hunt through thousands of satellite images, cross-reference water monitoring across decades, surface anomalies, connect disparate data sources. The closure engineer uses that synthesis to make the actual decision: the one that moves. The winners in AI-for-infrastructure aren't companies that replace expertise. They're companies that operationalize evidence synthesis so experts can spend their time on judgment, not paperwork. What we're doing We're running closure and remediation assessments across tier-1 mine operations in Canada. We're working with regulatory and federal-care closure teams on legacy sites. Our customers are in active discussions for multi-site rollouts across their operating portfolios. We're expanding that footprint and we need Implementation Specialists to own the deployment at each site. What we're looking for: Implementation Specialist You're a geotechnical engineer, closure specialist, or environmental engineer with 5+ years of site assessment or closure experience. Your expertise is rooted in one core area: whether that's geotechnical assessment, hydrology, closure operations, regulatory workflows, or environmental engineering. But you have hands-on closure experience across the problem space. And you're genuinely interested in how AI and agentic workflows can change how closure assessment gets done. You've worked with operators, regulators, or both on closure assessments and remediation planning. You understand what evidence matters to regulators, what moves risk engineers, what gets signed off on. You're comfortable with early-stage deployments (we're not at turn-key yet). You want to be in the field more than in the office. And you see a tool that could actually change how closure gets assessed and you want to own that transformation operationally. What you'll do: Initial intake: Work with site operators and engineers to understand closure scope, data landscape, and decision timeline Data synthesis: Coordinate with our data team to ingest site-specific sources (satellite, water, geology, operations records, regulatory filings) Workflow customization: Build closure-specific playbooks (artifact templates, citation structures, decision workflows) so TRAIL's synthesis matches how that site's team actually works Deployment to operations: Walk the closure team through the platform, integrate it into their review cycles, ensure the first assessment lands traceable and trusted Feedback loop: Sit with the team through their first closure decisions, capture what worked, feed findings back to product You're operationalizing a new way closure assessment happens: one site at a time. The offer: Salary: $120k–$150k CAD (negotiable based on experience) Equity: Early-stage grant (vesting over 4 years, 1-year cliff) Remote-first: Distributed team across Hal

KorrAI is looking for an InSAR Processing & R&D Engineer to work on specific high impact problems in our InSAR pipeline. This is not a generic GIS or remote sensing role. We are looking for someone who has worked close to the InSAR processing stack and understands how these systems behave in practice. You will work on focused technical scopes such as atmospheric correction, co registration, calibration, and interferogram quality. The goal is to take known gaps in our pipeline and turn them into working, production ready improvements. This role sits between research and engineering. You should be comfortable implementing methods, not just evaluating them. What this role is This is a scoped, part time contractor role focused on improving specific components of an existing InSAR pipeline. We already know the key problem areas. The work is to pick them up, solve them properly, and integrate improvements back into the system. You will not be building from scratch. You will be improving a production system where performance, accuracy, and scalability matter. These are independent, well defined technical scopes, not a single long project. Example areas of work Depending on fit and priority, scopes may include: Improving atmospheric correction methods across C band, L band, and X band data Investigating and improving co registration performance and diagnostics Enhancing interferogram quality control and error detection Improving calibration workflows using GNSS and validation sites Supporting unwrapping related R&D and quality assessment Testing and integrating methods from research into the pipeline Evaluating processing tradeoffs across sensors, DEMs, and stack configurations Contributing to scalable processing components for large area workflows What you’ll do Take ownership of a defined technical scope and propose an approach Review relevant literature and existing pipeline behavior Implement and test solutions in Python within a Linux environment Work with the team to validate outputs against real world data such as GNSS and reference sites Document results, limitations, and next steps Deliver work that can be integrated into the production pipeline What we’re looking for We are looking for someone who has worked hands on with InSAR processing and can operate close to the core pipeline. This role is focused on solving real technical problems in a production system, not just analyzing outputs. Hands on experience with InSAR processing including interferograms, co registration, atmospheric effects, unwrapping, and calibration Comfortable working close to the processing pipeline, not just GIS or visualization layers Ability to take a method or idea and implement it into a working solution Strong Python skills and comfort working in Linux environments Experience with Git and collaborative development workflows Ability to debug real world issues such as noise, coherence loss, and DEM errors Not a pure researcher or generic GIS profile. Looking for someone who can both understand the science and build systems Familiarity with SNAP, ISCE, StaMPS, Gamma, or similar tools Experience working with multiple SAR sensors such as Sentinel 1, TerraSAR X, COSMO SkyMed, or SAOCOM Experience with cloud or batch processing environments such as AWS or Docker Experience with validation approaches such as GNSS, corner reflectors, or ground truth data This role suits someone who enjoys solving technically challenging problems and can contribute quickly without heavy structure. Working style This role is best suited to someone who likes solving narrow, technically challenging problems and can work independently with minimal supervision. You should be comfortable working with partial context and figuring things out without full specifications. Engagement structure Part time contractor role up to around 20 hours per week Initial engagement of up to 6-12 months with potential extension Work is structured around specific technical scopes Remote To ap

Why we're building TRAIL We started KorrAI to monitor ground deformation on critical infrastructure using satellite data. Millimeter-scale precision on where the earth moves. But after working with operators and engineers, we hit the real constraint: synthesis. A closure engineer inherits decades of scattered data. Satellite imagery, water monitoring, geology reports, soil surveys, historical records, regulatory filings. They spend weeks manually reconstructing site history before they can assess remediation options. Regulators re-ask the same questions because there's no traceable record of what was checked and why. The bottleneck isn't data availability. It's operationalizing all of it: building a workspace where closure teams can ingest multi-modal sources, surface what matters, and produce reports that cite everything. We call it TRAIL. The thesis Billion-dollar infrastructure decisions rest on scattered evidence. Mining closure, construction risk, data center siting are high-stakes, high-evidence environments. Being wrong costs regulatory delays, remediation overruns, or worse. AI won't replace closure engineers or regulators. But AI can do what humans shouldn't: hunt through thousands of satellite images, cross-reference water monitoring across decades, surface anomalies, connect disparate data sources. The closure engineer uses that synthesis to make the actual decision: the one that moves. The winners in AI-for-infrastructure aren't companies that replace expertise. They're companies that operationalize evidence synthesis so experts can spend their time on judgment, not paperwork. What we're doing We're running closure and remediation assessments across tier-1 mine operations in Canada. We're working with regulatory and federal-care closure teams on legacy sites. Our customers are in active discussions for multi-site rollouts across their operating portfolios. We're expanding that footprint and we need Technical Advisors to shape the product as we scale. What we're looking for: Technical Advisor We want a technical advisor who: Has spent significant time in site closure in one core area (geology, hydrology, operations, regulatory, geotechnical) and a genuine passion for how technology can solve closure problems Understands the data landscape: what exists, what's missing, what's hard to access Knows how decisions get made and what evidence moves them Can tell us when we're solving the right problem and when we're solving theater This isn't a consulting retainer. We want advisors embedded enough to influence product direction, but independent enough to be credible to the market. What you'll do: Advise on closure-specific playbooks and workflows (what data matters, how it gets used, what gets signed off on) Review product decisions and early deployments; flag what doesn't match the real problem Connect us to the closure community: operators, regulators, engineers who need this Lend credibility and judgment as we expand operationally The offer: Cadence: weekly advisor calls + async feedback on key product decisions. Not a full-time commitment. You're advising us because you've seen closure problems firsthand and you want to shape how AI gets embedded into the solution not because you're a tech expert. Term: 12-month initial term, renewable Logistics: Distributed team (Halifax, Toronto, remote). You stay wherever you are. Calls scheduled around your timezone. Before you reach out: We want to understand your perspective on the problem. If you're interested, send a short response (a paragraph or two) to these questions: What's the biggest bottleneck you've seen in how closure assessments actually get made? Think about the data, the decisions, the timeline, the back-and-forth with regulators. What slows things down most? If you could offload one part of closure assessment to an AI agent, something that currently eats weeks or months, what would it be? And what would that actually solve for you operationally? We're not