The hiring map
KoBold has drilled 140,000 meters of crust across four continents with a team that has swollen past 600. The company treats hiring as a geological problem: find the right formation, drill the right holes, don't stop until the resource is defined.
The functional map splits in two. On one side sit the remote-first technology teams — data scientists, software engineers across the stack, and engineering managers for data systems — who build the AI platform that ingests geophysical, geochemical, and geological data to predict where copper, cobalt, lithium, and nickel actually live.
On the other side sit the field operations teams: exploration geologists, drilling supervisors, metallurgists, permitting specialists, and the community-engagement managers who negotiate the social license that lets the drills turn.
Between those poles sits a thin but critical layer: the integration roles that translate model outputs into drill targets and drill results back into model updates. The company's public remarks emphasize "brilliant people" and "deserving people" who get a shot regardless of whether they have a direct line to leadership. That philosophy shows up in the remote postings (global, no relocation required) and in the Zambia hiring push that prioritizes local talent alongside expatriate specialists.
The 2026 work program — six major objectives including strategic procurement decisions that will demand dozers, machinery, and labor at industrial scale — will test whether the two hiring engines can feed each other fast enough.
Compensation architecture
KoBold's pay structure sits at the intersection of mineral exploration and advanced computing, two sectors that historically operate on different scales.
Zero G Talent's job board data, drawn from 23 salaried roles across engineering, science, and leadership, shows a base salary band spanning $90,000 to $256,000 per year with a median of $200,000. That range reflects the breadth of roles required to run an AI-driven exploration company that is simultaneously building a $2 billion copper mine in Zambia.
Zero G Talent reported that at the top of the posted range, a Vice President of Business Development covering three continents carries a band of $250,000–$350,000. The role sits at the nexus of project finance, government relations, and offtake negotiations: the commercial engine behind the Mingomba development.
Zero G Talent's data shows that legal leadership follows closely: an Assistant General Counsel, Employment (remote, Eastern Time Zone) is banded at $210,000–$265,000, reflecting the regulatory complexity of operating across U.S., Canadian, and Zambian jurisdictions.
Zero G Talent found that a Data Scientist (Global, Remote) spans $140,000–$260,000, the widest band in the dataset, signaling that KoBold prices data talent by demonstrated impact on exploration outcomes rather than years of experience.
Zero G Talent's figures show Software Engineer (All Levels, Remote) runs $150,000–$240,000, while according to Zero G Talent, the Frontend Software Engineer (Remote) band is $160,000–$240,000. The Engineering Manager, Data Systems (Remote) sits at $200,000–$240,000, a narrower band consistent with a defined scope: leading the infrastructure that ingests geophysical, geochemical, and drilling data into KoBold's proprietary models.
| Role | Location | Base Salary Band (USD/year) |
|---|---|---|
| Vice President of Business Development | North America, Europe, Africa | $250,000 – $350,000 |
| Assistant General Counsel, Employment | Remote (ET) | $210,000 – $265,000 |
| Data Scientist (Global) | Remote | $140,000 – $260,000 |
| Frontend Software Engineer | Remote | $160,000 – $240,000 |
| Software Engineer (All Levels) | Remote | $150,000 – $240,000 |
| Engineering Manager, Data Systems | Remote | $200,000 – $240,000 |
The board data does not break out equity grants, refresh cadences, or benefit details by role. Publicly, KoBold has raised over $500 million from investors including Jeff Bezos, Bill Gates, Mark Zuckerberg, Sam Altman, Marc Andreessen, Michael Bloomberg, Jack Ma, and Andreessen Horowitz; these capital structures typically support meaningful equity packages for early- and mid-stage employees.
Candidates should expect equity to be a material component of total compensation, particularly for roles at the engineering manager level and above, but the company does not publish grant sizes or vesting schedules in its postings.
Remote eligibility is explicit across the technical and legal roles listed; the VP Business Development role is the only one tied to a multi-region travel mandate. That distribution mirrors the company's operational reality: a distributed data and software team feeding a centralized exploration effort that now employs over 600 people and has drilled more than 140,000 meters at Mingomba in a three-year window, work that demands both cloud-native collaboration and periodic field deployment.
For candidates benchmarking offers, the median of $200,000 base is the anchor. A data scientist at the top of band ($260,000) earns roughly the same base as an engineering manager at the top of theirs ($240,000), but the VP role opens a clear step change above $300,000. Equity upside, while unpublished, scales with the company's trajectory from exploration to construction, a transition that historically revalues early technical equity substantially.
Inside the interview room
KoBold's hiring motion reflects its origin story: the founders "frontloaded the company not with mining experts but with data scientists and software engineers from places like Google, Apple and Meta," the Berkeleyside profile reported. That DNA still shapes who gets interviewed and how. Josh Goldman has described exploration as an information problem: "We don't drill for metals, we drill for information." The team builds "dozens of different products that work together" across three themes: data acquisition, probabilistic modeling, and decision support.
The live board shows active requisitions across a wide aperture: Vice President of Business Development spanning three continents, an Assistant General Counsel for employment law, data scientists, frontend and backend software engineers at all levels, and an engineering manager for data systems.
KoBold operates as a full-stack technology company that happens to run a minerals exploration portfolio: 60 projects on four continents, a $600 million Zambia commitment, and a proprietary hardware-software stack (Hyperpod, helicopter-borne EM, drone geophysics, and the analytics layer that ties them together).
That distribution suggests a process that weighs asynchronous collaboration skills heavily for the remote cohort, while the field-facing roles add a separate evaluation for cross-cultural operating stamina; Zambia alone employs roughly 300 people, mostly Zambian, under an in-country CEO.
The company's transparency about its technical architecture — helicopter EM, Hyperpod, the Zambia dewatering challenge — suggests they will answer concretely when candidates ask about interview stages, loop composition, and what a "strong yes" looks like at each gate.
Where the work lives
KoBold's workforce splits between a distributed remote core and a handful of physical hubs tied to its exploration and development pipeline. The company's own job board makes the remote tilt explicit: five of the six most recent postings (data scientist, frontend engineer, software engineer (all levels), engineering manager for data systems, and an assistant general counsel role) list "Remote" or "Remote (Eastern Time Zone)" as the location.
Only the vice president of business development carries a multi-region mandate (North America, Europe, Africa) without a remote tag, suggesting travel-heavy field engagement rather than a fixed desk.
The most concrete physical site in KoBold's portfolio today is the Mingomba project on Zambia's Copperbelt. The company has moved the asset into development work targeting first copper in the early 2030s, a timeline that aligns with Zambia's national goal of 3 million metric tons of annual output by 2031. Mingomba sits on sediment-hosted, stratiform copper in Roan Group rocks that can deliver grades well above global averages, but at depth, in water-bearing ground that demands shaft sinking, heavy dewatering, ventilation, and refrigeration.
Those are multi-year capital items. If Mingomba mirrors nearby deposits like Konkola and Mufulira, the operation will need extensive ground support, paste backfill for stope stability, and a mining method choice (bulk vs. selective) that trades off dilution against throughput. The early-2030s target implies a multi-year engineering and sinking program is already underway or imminent.
That program creates a second physical center of gravity: the Zambian site itself, where geotechnical drilling, metallurgical testwork, hydrogeological modeling, and infrastructure procurement (hoists, winders, refrigeration plant, transformers, mill components) converge. The nai500 analysis notes that credible schedules must show early orders for long-lead items and contracts locking in major equipment deliveries; the absence of which is a red flag.
Power and water shape feasibility as much as geology: Zambia's hydropower-dependent grid suffered load curtailments in 2024, and deep mines need reliable baseload for pumping and cooling. A binding power supply agreement with drought contingency, plus dedicated backup generation or firmed grid capacity, is a prerequisite. Logistics add another layer: Copperbelt mines benefit from established road and rail, but sinking consumables, cement for backfill, and refrigeration components move through congested corridors where shipping delays cascade.
Beyond Zambia, the research does not document KoBold's U.S. office footprint, lab space, test cells, machine shops, or integration floors. The a16z thesis argues that the winning model in critical minerals is a vertically integrated operator ("the factory is the product") and points to processing co-located with extraction as a source of feedback loops that improve recovery and quality control.
But whether KoBold maintains a domestic hardware prototyping facility, a core-logging lab, or a sensor-integration floor is not disclosed in the available sources. The remote-heavy hiring pattern for software and data roles suggests those functions operate distributed, while field geologists, mining engineers, and project development staff likely rotate through Zambia and any corporate headquarters the company maintains.
What's clear: the work happens where the ore body dictates. The remote roles build the digital stack — exploration targeting, data systems, modeling — that feeds the physical development program in Zambia.
The split is structural, not incidental. Candidates should expect the software and data teams to stay distributed, while engineering, geoscience, and project execution roles will demand significant field time in southern Africa as Mingomba advances from study to sinking.
The profile that lasts
The public record — news coverage, the company's own site, and third‑party profiles — focuses on KoBold's exploration technology, funding rounds, and partnership announcements rather than on the human‑capital patterns that predict tenure.
What follows is an inference drawn from the roles KoBold actually recruits for, the compensation structure it publishes, and the operational model those two facts imply. Treat it as a working hypothesis, not a documented finding.
The board data shows a consistent hiring profile: senior individual contributors and managers in data science, software engineering (frontend and full‑stack), business development, and employment law. Every one of the six recent postings is listed as remote, and the salary band clusters between $140 k and $350 k with a board‑wide median of $200 k. That distribution tells a story. A remote‑first organization that pays a $200 k median to data scientists and engineers is selecting for people who can deliver complex, ambiguous work without daily physical oversight. Self‑direction, asynchronous communication discipline, and the ability to structure one's own investigation pipeline become baseline requirements, not differentiators.
The technical roles (Data Scientist (Global), Frontend Software Engineer, Software Engineer (All Levels), Engineering Manager, Data Systems) share a common thread: they sit at the intersection of geoscience domain knowledge and modern data‑engineering tooling.
Candidates who thrive in that intersection tend to exhibit two complementary mindsets. First, a tolerance for dirty, sparse, and non‑stationary data (the kind that comes from drill logs, satellite spectroscopy, and historical geological maps) and the creativity to extract signal where standard pipelines fail.
Second, a willingness to build and maintain the infrastructure that makes that extraction repeatable: versioned notebooks, reproducible environments, automated QA on ingest. The job titles alone don't capture this; the salary spread ($140 k–$260 k for individual contributors, $200 k–$240 k for the engineering manager) signals that KoBold values the engineering rigor as much as the statistical insight.
The business‑development and legal roles (Vice President of Business Development (North America, Europe, Africa) at $250 k–$350 k and Assistant General Counsel, Employment at $210 k–$265 k) reveal a second success vector: comfort operating across jurisdictions, cultures, and regulatory regimes while negotiating partnerships that involve sovereign resource ownership, community relations, and long‑horizon capital.
The VP role's geographic scope implies frequent context‑switching between legal frameworks in Canada, Zambia, the DRC, and the U.S., often without a local office. The employment counsel role, though remote and U.S.‑focused, sits inside a company that hires globally; the person in that seat must anticipate how labor law, equity treatment, and data‑privacy rules interact across the same footprint. Success here correlates with a lawyer's or deal‑maker's ability to translate technical risk into commercial terms that a mining minister, a community chief, and a New York‑based LP can all understand.
Equity is part of every offer, but the board data does not publish grant sizes, vesting schedules, or refresh policies. In the absence of those specifics, the compensation architecture (wide bands, high median, remote‑first) suggests a culture that rewards outcome ownership over presence.
People who negotiate for equity upside, track their own contribution to milestone‑based value inflections (a new deposit model validated, a partnership signed, a pipeline automated), and can articulate that contribution in a written memo will likely out‑perform peers who expect recognition to flow from meetings attended.
No public source identifies a "KoBold archetype": no founder blog post on hiring philosophy, no Glassdoor pattern, no alumni newsletter.
The closest proxy is the company's own description of its work: "AI‑guided mineral exploration" that fuses "geoscience, data science, and operations." If that fusion is the daily reality, then the employees who stay are the ones who treat the boundaries between those disciplines as seams to be engineered, not walls to be defended. They write code that a geologist can audit; they design surveys that a model can ingest; they negotiate access agreements that a data pipeline can track. The research cannot confirm this. The hiring pattern, however, makes it a reasonable bet.
KoBold began with a handful of people in a room. Now the drills turn at Mingomba, 140,000 meters deep, and the team that put them there spans continents. The next 600 hires will not be found by posting a requisition; they will be located the same way the copper was: by mapping the formation, drilling the right holes, and trusting the signal when it returns.
Working in frontier tech? Zero G Talent tracks the openings: see every open KoBold Metals role, browse frontier tech jobs, the companies hiring, and the people building the field.