The Two Roles and the Company Behind Them
Two Pennington, New Jersey, openings listed for Village on Zero G Talent’s live board carry teaching titles, while the board’s company record names the founders of Brooklyn software company Village Labs. Both were added in the past seven days. That leaves candidates with a research problem rather than a hiring rubric: they can compare dated copy and peer screens, but no first-party source verifies the criteria behind these openings.
| Opening | Location | Listed pay |
|---|---|---|
| Lead Teacher - The Village | Pennington, NJ | $17.25–$20.50 an hour |
| Assistant Teacher - The Village | Pennington, NJ | $16.25–$17 an hour |
The title order is the only hierarchy visible in those two listings. The company’s own materials supply the context the listings omit.
The company’s About page provides the fuller account. Founded in 2023 and based in Brooklyn, Village Labs builds AI software for employee stock ownership plans, or ESOPs. Its flagship product, Peninsula, gives advisory firms a shared place for client data, institutional knowledge and recurring workflows. That identifies Village Labs as applied AI and financial software for ESOPs.
An August 30, 2026 profile from The Rundown said the company had moved away from its earlier, broad “company brain” description to a narrower focus on AI and financial systems for employee ownership plans. The shift ties the products to this market: they now cover repurchase-obligation forecasting, trust and cash-flow planning, document analysis and conversational access to specialist workflows.
In the company’s product documentation, Peninsula agents work inside the documents and spreadsheets advisers already use. The same documentation lists sustainability and trust accounting under “In Development,” and third-party administration, distributions processing and transaction management under “Roadmap.” Village Labs says clients operate in isolated environments, their data does not train Village Labs’ models, and role-based permissions and audit trails record every action and access. The company also says Peninsula supports professional work rather than replacing the fiduciary, legal, financial or administrative judgment carried by the people responsible for an ESOP.
NCEO’s analysis of U.S. Department of Labor Form 5500 research files reports that U.S. ESOPs hold $2.06 trillion in plan assets and include 15.1 million total participants, using 2023 data updated in January 2026. Those figures size the employee-ownership system, not Village Labs itself: they do not reveal its customer count, revenue or product adoption. They indicate the size of the market Village Labs serves without establishing its commercial position.
The founders divide product and technical work. Co-founder and Chief Executive Officer Roger Cawdette leads product development and customer relationships. He previously co-founded Finary and studied Technological and Social Innovation at Harvard. Co-founder and Chief Technology Officer Turja Chowdhury leads engineering, technical architecture, product delivery and platform reliability across Village Labs and Peninsula. He previously developed systems for enterprise banking, crypto accounting and IT services firms. Village Labs is Y Combinator-backed and keeps product, engineering, client-experience and repurchase expertise in Brooklyn.
The stated roles put customer relationships under Cawdette and broad technical responsibilities under Chowdhury. They do not identify who will review the teacher applications, and the board’s teacher openings name neither founder as a hiring manager.
The openings still do not connect that identity to Peninsula, ESOP work or a technical hiring track. That distinction governs the rest of the public record.
What the Job Posts Actually Screen For
The live board records for the Pennington openings name the roles but do not include the job text; older Village Labs records contain explicit terms for different work. The Village board and those dated records do not explain the relationship between the two sets of listings.
What the live records prove
The live records show the title, location and pay; the board also says the two roles were added in the past seven days. They do not include the job description, formal qualifications, application instructions, interview stages or applicant-tracking-system rules.
For a candidate, the defensible keyword set is therefore the exact title and location. The records cannot establish whether Village requires a degree, teaching certification, classroom experience, a background check, lesson-planning work or a particular software tool.
Salary may affect whether an applicant chooses to apply, yet no first-party source identifies compensation as a selection criterion.
The live record shows what Village is advertising, not what criteria Village applies when selecting applicants.
Attribution in a company profile does not turn third-party job copy into a first-party statement of Village’s current hiring criteria. That makes the published filter an absence, not a decodable sequence of checks.
What the dated Village Labs copy says
Older records reveal the language that once surrounded Village Labs hiring, but they cannot describe the live openings. An April 22, 2024 third-party company record for a Solutions Architect role tied to customer onboarding asked for at least two years of experience building customer-onboarding and setup tools, at least three years of full-stack development, and a track record deploying and maintaining complex systems at scale. It also expected direct client work on technical requirements and business goals. React, Next.js and TypeScript appeared as preferences rather than requirements, while the role covered the processes, systems and team needed to grow Village from zero to 100 customer onboardings per week.
That description frames a specific operating problem: build onboarding capacity while the company is still early. It combines coding with client work and treats the named technologies as preferences. The weekly onboarding target signals the intended scale, not proof that Village will screen every applicant for prior onboarding experience elsewhere.
A Wellfound record for a senior software engineer gave conflicting experience requirements: one floor at four years and another range extending beyond eight. The mismatch weakens any attempt to use the range as a clean threshold. The role also asked candidates to take new products from zero to one, move from system design through deployment, use AI to improve development velocity, work across AI systems and traditional application code, and operate autonomously.
The stack language was broad: “Any languages are fine,” with Go, TypeScript and Python named as the company’s primary languages. The application note was direct: “We encourage you to drop us a line even if you feel over- or under-qualified.” That is direct wording from the dated record; it is not a criterion for the live teaching openings.
Signals are not a published screen
The older copy points toward candidates who can take ambiguous, early-stage work from zero to one. References to onboarding, client goals and systems that scale suggest ownership beyond a narrow coding specialty. The emphasis on AI-assisted velocity, plus a reference to getting roughly five times as many model tokens for the same spend, also points to cost control in day-to-day work.
The culture language is harder to test. Terms such as “obsessed with making customers happy” and “do whatever it takes” describe intensity rather than a verifiable credential. They may shape an interview, but the supplied sources do not identify the interviewers, questions or scoring.
The dated software copy and live teaching posts cannot be merged into one rubric: one describes engineering work from another time; the other names jobs without explaining them. The next useful evidence would be the full live descriptions or a first-party hiring statement from Village Labs.
Candidate Moves in Response to an Undocumented Screen
Without a published rubric, a candidate’s advantage comes from making each application testable, not pretending the company has one. The Village company record and the broader research provide no applicant log, outreach log, source-of-hire field, Village-specific rejection reason, applicant response rate or documented hiring outcome. Those are the fields needed to compare tactics; without them, confidence would be manufactured.
Public posts on Village Labs’ own LinkedIn account do not fill that gap. A 2026 post described the conversation around artificial intelligence moving from “What is this?” to “How can I use this?” and said the company had given two “AI-Native” talks totaling four hours. A later post focused on companies shifting from trying ChatGPT to improving how they operate. Neither post is hiring guidance, and neither asks for portfolios, work samples, referrals, essays or public demonstrations. The research includes no candidate account showing how the company responded to such work.
Referral evidence is thinner still. No supplied source documents a referral to Village Labs, a referral bonus, or an applicant asking an employee for a recommendation. The nearest example comes from HRKatha’s reporting on recruitment through Indian village and WhatsApp networks. Naxatra Labs co-founder and CEO Abhilash Maurya printed Gujarati job descriptions, circulated them through local WhatsApp groups, and offered Rs 200 if the referred person stayed two weeks. A trusted contact carried information a job portal could not:
“This worked because trust was embedded in the chain. Whatever due diligence HR would normally perform was already being done informally by the person recommending the candidate.”
Emmanuel David, a senior HR leader, said. The report warns that community hiring favors people with strong networks and can weaken objective evaluation. It does not show that Village Labs uses referrals, pays referral bonuses, or treats a recommender’s endorsement as a substitute for review.
The Anti Job Board gives a detailed process, but its own limits narrow the reach. Its Village Labs page describes a roughly weeklong, three-stage process, says Greenhouse handles applications while no dedicated recruiters do. It says a founder or department head would usually make the decision, while explicitly warning that the company has not confirmed this. The page advises applicants to tailor their résumé, apply promptly, send a follow-up and use direct outreach. It also declares: “Direct outreach wins.”
These are tactics, not observed results. The page’s claim that a follow-up roughly doubles reply rates has no published sample, comparison group or denominator, and it names no example showing that a direct message produced a reply at Village Labs. A small company may benefit from direct contact, especially when founders review applications between other duties, but that remains a hypothesis rather than a hiring rule. Until Village confirms who reads the inbox and what each stage tests, candidates should treat outreach as a variable in a test, not a proven route past the screen.
The page also reports no live vacancies, while Wellfound’s October 2026 page listed six jobs and Zero G’s board showed two. The pages do not establish a synchronized inventory, and they do not identify the people behind the listings. The research cannot support a conclusion that referrals outperform direct outreach, early applications win more often, or public work raises interview odds.
Candidates can still turn the uncertainty into usable evidence. Before applying, open a dated ledger: record the role and posting time, then note the tailored materials, referral, direct message, follow-up, stage reached and outcome. Later, the ledger will show which action changed the result—or whether none did.
If a full role description appears, attach a focused artifact tied to its stated problem rather than a generic portfolio. Until then, do not present public work as a published requirement. For outreach, identify the exact opening and ask one specific question; do not substitute candidate ambition for missing job details. For a referral, ask the recommender for useful context and ask Village whether referrals affect review. Treat Roger Cawdette and Turja Chowdhury as people to ask, not confirmed decision-makers.
A fast reply is not a hiring rule, and a referral is a route, not a verdict. The alternative is to examine screens that make their tests visible.
Peer-Company Screens as Evidence, Not Confirmation
TechCrunch reported that 4,500 people signed up for Stripe’s 2012 capture-the-flag contest in its first six hours. A 2026 Built In history of technical hiring says Stripe has hired programmers who discovered and exploited vulnerabilities in mock web apps. The Village board still supplies no comparable exercise. The closest current peer records come from defense software, hardware and federal technology, not teaching. They can identify selection methods without predicting Village’s questions or scoring.
What peer-company screens can establish
A recent Aced account, based on candidate reports, describes three stages for most roles and a practical evaluation at the technical stage. During the phone screen, software candidates usually solve coding problems live on HackerRank; in the onsite, they work in a shared editor. A BMH Learning overview separately says hardware candidates may discuss engineering scenarios. The Aced account’s candidate reports mention no take-home assignment. Rather than rely only on résumé claims, the process includes observable work. The transferable question is precise: does Village use live exercises, and what does it score?
The same account places “Why Anduril?” in both the recruiter screen and a final behavioral interview with a senior leader, where it tests comfort with defense work. Hiring also varies by team: a recruiter may circulate a candidate’s profile to another team if the fit is better there. A technical screen and mission screen can operate separately within one company. The defense-specific question does not establish a Village criterion.
The U.S. Tech Force’s USAJOBS notice, open from October 13, 2026, offers a published, gate-based example. The federal program does not require a specific degree, but it requires GS-14 general and technical competencies plus specialized experience. Candidates move through USA Hire assessments, a General Coding Assessment and an automated digital interview built around predetermined rubrics. For the coding assessment, the notice says:
“You must meet or exceed the passing score to be considered.”
The notice presents the cut score as an additional gate alongside the competency list. U.S. Tech Force is a two-year federal staffing program. Appointments last no more than two years, can extend to four, and give no automatic right to continued federal employment. Those are features of the federal program, not evidence of Village’s process.
Anduril’s Grand Prix offers a different route. A February 2026 Built In report records a $500,000 prize pool behind a drone-racing contest; the highest scorer could receive a direct interview with a hiring manager. It accepted individuals and teams of up to eight without requiring professional credentials or certifications. Competitors could not modify the drones, so performance depended on the speed and precision of their algorithms. A separate employment-eligibility check could block a winner unable to obtain and maintain an active security clearance; that candidate would receive $10,000 instead of a job.
The structure matters more than drone racing: it links demonstrated performance to a hiring conversation, then applies a nontechnical check. Its value here is structural rather than predictive.
The longer record points in the same direction, with clear date limits. The same 2026 Built In account lists Meta as hosting its Hacker Cup since 2011, Google as holding coding contests until 2023, and Yelp and Walmart as recruiting data scientists through Kaggle competitions. Those are dated precedents, not current peer screens. Together, they support a narrow claim: some employers have hired from visible performance events instead of conventional background signals.
What peer-company screens cannot establish
A defense contract can be public without making a hiring rubric public. An October 2026 report said Arlington, Virginia-based Shift5 joined the Army’s Next Generation Command and Control effort in February 2026 through a partnership with Anduril, integrated its technology with Anduril’s Lattice software, and took part in exercises including Ivy Sting and Project Convergence. Shift5 also secured a five-year Army contract with a maximum potential value of $99 million, with funding determined through individual orders. Those facts show what the company works on. The research documents no Shift5 interview rubric, so the contract cannot reveal what candidates are rewarded for.
Mission preferences are equally difficult to transfer. An August 21, 2026 GeekWire profile said Anduril’s Seattle-area pitch emphasized “grit,” ownership of difficult engineering problems and willingness to test work in the field. It also described an in-office culture partly because the company builds hardware. Those conditions describe Anduril, not a universal frontier-tech standard. No company source says Village wants defense experience, a clearance or a drone-racing record, and the supplied Village materials do not publish an in-office schedule or technical test.
Separate observation from transfer. When a peer source assesses live work, applies a scored threshold, asks a repeated mission question or imposes an eligibility check, ask whether that work resembles the Village opening. An Anduril coding exercise tells a teacher candidate little about teaching. A peer work sample can suggest what evidence to request, but it cannot establish how Village judges one.
Before applying, a candidate can ask the employer behind the Village listing for the assessment format, rubric, stage count, live-work expectations and recurring mission questions. A direct answer would replace a peer-based hypothesis with company evidence.
The missing job description is not a code to break. Until Village supplies the full posts or answers those questions, the public record still ends where it began: two teaching openings, the profile of a software company and no published filter. A peer contest may sharpen the question; it cannot become the answer.
Working in frontier tech? Zero G Talent tracks the openings: see every open Village role, browse frontier tech jobs, the companies hiring, and the people building the field.