Who Gets Hired and Onto Which Teams
The team at Solve Intelligence is small on paper and deep on credentials. A 2025 TechCrunch interview with co-founder and CEO Chris Parsonson counted 15 employees; the Y Combinator listing now pegs headcount at 50, roughly triple in a year. Parsonson has framed the hiring bar in unusually plain terms: a "lean, highly technical team of AI PhDs and software dev experts capable of building the features our users want faster than our competitors," working alongside patent attorneys who can translate between legal practice and machine-learning output.
That three-part composition (attorney, AI researcher, engineer) is the spine of who gets hired.
The three teams and what they actually do
Patent attorneys at Solve are not generalist legal hires. The company targets practitioners who understand the specific shape of patent work: drafting applications, responding to office actions, building claim charts across competitor products. Solve's homepage reports that 30% of its customers come from Vault 30 IP Law Firms, 21% from AM Law 100 firms, and 20% from Chambers Band 1–4 IP firms, so the in-house attorney team needs enough range to support that mix, from a partner at DLA Piper comparing a hundred patents against six products to a sole practitioner drafting a single provisional. The company's current career page lists an opening for a Patent Litigator in London, Munich, or New York in the $150,000–$250,000 band, which signals that legal hires operate close to the customer rather than in a back-office review function.
The AI research side is what separates Solve from competitors like PatSnap, IPRally, Harvey AI, Casetext, and Robin AI, none of which focus on the patent drafting pipeline end-to-end. Researchers here work on retrieval-augmented generation pipelines, multi-step LLM orchestration, and patent-specific customization, including the "ability to customize the AI to a user's own unique style, and then switch between different styles for different jurisdictions, technology fields, and clients" that Parsonson has called Solve's most popular feature. Posted AI Engineer roles on our board sit in the $100,000–$250,000 band out of London, consistent with a startup that wants senior applied researchers who can ship to production rather than publish.
Software engineers round out the trio. Live listings on our board include Full-Stack (Back-End Leaning) and Infrastructure Engineer roles in London, both in the $100,000–$250,000 range; the Infrastructure role tops out closer to $250,000, suggesting backend and platform work carries a small premium. Solve's engineering floor looks weighted toward the systems that sit underneath the model layer: proprietary algorithms, sequential LLM calls, RAG pipelines, and the secure enterprise-AWS hosting that addresses attorney confidentiality concerns.
Background, not just credentials
Solve's public framing emphasizes demonstrable skill over pedigree. That dual fluency shows up in the employee mix the company describes: people who can read both a transformer architecture diagram and a patent claim tree. TechCrunch reports that Solve has seen 25% month-on-month revenue growth since launch, profitability before its Series A, and a customer base that crossed 700 IP teams across six continents, with users producing more than 433,000 patent applications and 103,000 office-action responses on the platform.
The Series A from Microsoft and the Series B announced in late 2025, bringing total backing to $55 million from Y Combinator, Microsoft, Thomson Reuters, 20VC, Operator Collective, Visionaries Club, and others, fund the next hiring wave. Parsonson has said the capital goes toward R&D in life sciences, freedom-to-operate analysis, claim-chart licensing and litigation, and portfolio tools for in-house IP teams. Each of those product lines needs the same composition: attorneys who know the workflow, researchers who can extend the model layer, engineers who can ship it.
What It Pays
Solve Intelligence publishes wide salary bands on its open roles, and the spreads look more like venture-scale options than tight corporate ranges. The table below summarizes the company's live postings on Zero G Talent's job board:
| Role | Location | Band (USD) |
|---|---|---|
| AI Engineer | London | $100,000–$250,000 |
| Full-Stack Engineer | London | $100,000–$250,000 |
| Full-Stack (Back-End Leaning) | London | $100,000–$250,000 |
| Infrastructure Engineer | London | $120,000–$250,000 |
| Patent Litigator | London / Munich / New York | $150,000–$250,000 |
Across 17 salaried roles on the board, the full band runs $71,000–$250,000 with a median of $220,000 — a median that sits at the top of the listed spread rather than its midpoint, which signals the lower end of the band is closer to a floor for less-senior hires than to a typical offer.
Third-party aggregators track the same company at lower medians, and the gap is worth reading. Levels.fyi puts the median Solve Intelligence salary at $140,169 for a Software Engineer, while jobs-radar.com reports a $160,000 median across 15 self-reported pay disclosures, with most offers falling between $122,500 and $175,000. salaryguide.com gives a typical range of $90,000–$150,000. Those numbers almost certainly reflect earlier-stage hires and a narrower slice of the workforce than the company's current headcount, which salaryguide.com reports as up 94.7% year over year. The 15 open roles currently listed on LinkedIn span London, New York, and Munich, and the spread on the company's own postings lines up with a team that is actively hiring senior engineers and litigators, not backfilling junior seats.
On equity, the company has not published specific option grants or strike prices in the research available, so the picture stays qualitative. TechCrunch reported that Solve raised $12 million in Series A funding led by 20VC in April 2025, bringing total capital raised at that point to $15 million; the company has since raised additional capital to a reported $55 million total to build out AI patent tools, and Parsonson told TechCrunch the startup was already profitable before the Series A and had more in the bank than its $3 million seed. A profitable two-year-old AI company with multiple funding rounds behind it and backers including Microsoft typically grants meaningful equity packages, but no public numbers confirm what that means in dollar terms for individual hires. Candidates should treat equity as part of the total compensation conversation rather than a known figure.
Benefits and perks are similarly thin in the public record. The research confirms three physical sites (London headquarters, plus New York and Munich offices) but does not detail health coverage, retirement matching, parental leave, or remote-work policy. Solve Intelligence's careers page frames the work around the patent system itself ("The patent process hasn't changed in decades. We're making every step of it faster, cheaper, and more accessible"), which suggests the company's pitch to candidates leans on mission and the unusual intersection of legal domain expertise with frontier AI work rather than on a publicized benefits package. Anyone evaluating an offer should ask directly about equity grant size, vesting schedule, and benefits; the public data stops at the salary band.
What the Hiring Process Looks Like
Solve Intelligence runs a typical structured interview loop, and candidates should expect to clear several distinct stages before an offer lands: a recruiter screen, one or more technical evaluations, culture-fit conversations, and final validation, culminating in reference checks and onboarding.
Candidates who have made it through describe a process that leans toward the difficult end of the spectrum. Glassdoor users rated their interview experience there at just 20% positive, with a difficulty score of 3.4 out of 5. That ratio puts the bar in line with elite AI labs, where the volume of applicants and the technical specificity of the roles filter aggressively.
The stages
The first conversation is almost always a recruiter screen. This is where the candidate's background is verified against the role's must-haves, and where recruiters look for the signal that the person can actually do the job, not just present a polished version of one. With the rise of generative tools, this first round has become the most compromised step in modern hiring. Harvard Business Review's June 2026 reporting found that suspicious patterns appeared in roughly 60% of recorded first-round screening sessions for new-graduate software engineering roles, and close to 40% for mid-level technical roles, versus under 10% for less technical positions like account executive. For a technical AI shop like Solve Intelligence, that suspicion rate is the operating environment.
After the screen, candidates typically move into a deeper technical round. This is where structured interviews earn their keep: a 2023 reassessment in Industrial and Organizational Psychology found that structured interviews built around job-relevant questions, consistent scoring, and evidence of reasoning remain among the better predictors of actual job performance, while several traditional credential-based signals (pedigree, GPA, brand-name employers) turned out to be weaker than long-cited estimates suggested.
What gets candidates through
The candidates who clear the loop tend to share a few habits. Roughly one in four candidates screened in HBR's 2026 research outperformed what interviewers expected based on their résumé when the interviewer used a dynamic and unscripted style, meaning they could defend a counterintuitive tradeoff or think through a problem live rather than recite a polished answer. That's the behavior Solve Intelligence's loop is built to surface.
Practically, that means candidates should expect interviewers to introduce sudden constraints, ask them to evaluate flawed AI-generated artifacts, or push them to defend an unexpected decision. When candidates think out loud, admit uncertainty, and work through problems collaboratively, the incentive to lean on a generative tool drops, and the candidate's actual reasoning becomes visible.
What gets candidates eliminated
The disqualifiers cut both directions. The obvious one is dishonesty: the recent Cluely case, where 21-year-old Chungin "Roy" Lee raised $5.3 million in seed funding from Abstract Ventures and Susa Ventures for a tool designed to cheat on engineering interviews, illustrates how aggressively some candidates will try to game the loop, and how seriously employers have started treating the threat. Amazon told TechCrunch that its candidates must acknowledge they won't use unauthorized tools during the interview process; other firms are reintroducing in-person rounds for high-leverage hires.
The less obvious disqualifier is a résumé that reads as too perfect. Candidates have learned to stuff résumés with role-specific keywords, hide terms in white text to boost automated match scores, and increasingly draft with the same AI tools recruiters use. A 2025 study by researchers at the University of Maryland, Ohio State, and the National University of Singapore found that in simulated pipelines, candidates whose résumés matched the evaluating model's style were 23% to 60% more likely to be shortlisted than equally qualified applicants who wrote their own. For Solve Intelligence, that means the strongest applications are concrete: shipped projects, named systems, specific results. With 17 salaried roles on the board and a typical band of $71k–$250k (median $220k), the company can afford to be selective; the loop filters for the candidates whose résumés reflect what they have actually built.
Gartner anticipates that by 2028, one in four candidate profiles will be partially or entirely fake, a number that should sharpen any serious candidate's focus on demonstrating real, verifiable work rather than a manufactured presentation of it.
Where the Work Happens
Solve Intelligence runs its engineering work out of a single anchor site in London. Every active posting on the Zero G Talent board (AI Engineer, Full-Stack Engineer (Back-End Leaning), Infrastructure Engineer, Patent Litigator) lists London, England, GB as the work location, with the patent role additionally opened to Munich and New York. That concentration tells you how the company scales: there is one physical place where the team sits, prototypes, and ships, and remote hires orbit it rather than replace it.
The London base matters more than a typical office address would, because Solve Intelligence builds tooling that has to survive contact with real patent workflows, the kind of work where a wrong claim chart or a hallucinated prior-art reference costs a client a filing deadline. Engineers iterate against live attorney feedback, not just internal mocks. Proximity to clients and to in-house patent expertise is part of the engineering loop, not a perk.
For candidates, the practical implication is relocation. London hire, on-site or hybrid, is the default; the board carries no "remote, UK" or "remote, EU" filter for these roles. The Patent Litigator posting is the lone exception that explicitly opens Munich and New York, which suggests the company tolerates geographic spread only where licensed-jurisdiction workforces require it.
The lab footprint is small and deliberately so. Solve Intelligence is a software company; it does not run chemical reactors, ballistic test cells, or DNA processing lines, so it does not need the heavy infrastructure that anchors some peer AI firms. What it does need is secure compute, fast iteration on language-model pipelines, and close collaboration with the patent team that validates outputs. The London office is sized for that: cross-functional pods of AI engineers, full-stack engineers, and infrastructure engineers sitting near the litigators and patent professionals who grade their work.
London itself is a constraint and an asset. As a hub, it gives Solve Intelligence access to a deep pool of machine-learning engineers with NLP specialization, and to the European patent profession that the company serves. As a cost base, it is not cheap; the salary bands on the board reflect London market rates rather than discounted remote-economy pay.
For applicants evaluating the role, the takeaway is concrete. Expect to be in a London office, working in tight iteration loops with attorneys and litigators, building tools that have to clear a high accuracy bar before they ever reach a client.
Who Thrives at Solve Intelligence
The traits that correlate with success here are the traits that matter when shipping production AI under constraints, and the public record on AI hiring in 2026, combined with the company's own role mix, points clearly at who fits.
Three capabilities show up across the research as the differentiators for AI-native engineering and scientific work. Third, product-mindedness under tight constraints, because every role posted on Zero G Talent's board sits in a shipping path where someone has to decide what to cut, what to keep, and what to ship next.
Several behaviors recur in the public record of who does well in exactly this kind of seat. People who treat AI as a capability to integrate, not a replacement to manage, fit: Wharton professor Ethan Mollick's framing is direct: "the source of any real advantage in AI will come from the expertise of their employees, which is needed to unlock the expertise latent in AI."
They're less engaged and fulfilled. They feel less autonomy, purpose, and belonging." That cuts against the "ship under constraints" culture at any fast-moving AI shop, including Solve Intelligence, and is something candidates should weigh directly.
The practical read for applicants: the people who thrive at Solve Intelligence are the ones who can demonstrate, in the structured interview loop the company runs, that they ship AI systems to real users, that they treat the model as a teammate with sharp edges, and that they can name what they would cut to make a deadline. Anyone whose answer to "what should we ship next" is a feature list rather than a defended judgment call is signaling the wrong trait.
Working in frontier tech? Zero G Talent tracks the openings: see every open Solve Intelligence role, browse frontier tech jobs, the companies hiring, and the people building the field.