Vetto

Role Description We are opening a new opportunity for the Safety Project — Election Neutrality, a project focused on evaluating how AI systems behave in politically sensitive conversations. The goal of this project is to assess whether AI can remain neutral, balanced, and responsible during multi-turn interactions with users who express different political opinions, positions, and perspectives. For this project, we are looking for people who regularly engage in political discussions — whether at the dinner table, with friends, at work, online, or in other real-life contexts — and who can draw from their own debate experience to create realistic political interactions with AI. The project will involve scenarios related to international political discussions, where contributors will role-play users with specific perspectives and evaluate how the AI responds. Main Responsibilities - Understand the political context and perspective you are expected to role-play - Conduct a realistic multi-turn conversation with the AI - Defend or express the assigned political position naturally, as a real person would - Justify your evaluation using specific turns and evidence from the interaction - Identify whether the AI remained neutral, balanced, and responsible when responding to political opinions under pushback What We Are Looking For - Regularly engage in international political debates or discussions in real life - Have experience defending a political position under disagreement or pushback - Are interested in international politics, public debate, political discourse, or current affairs - Can follow detailed instructions and long guidelines carefully - Can evaluate AI responses objectively, even when the topic involves political disagreement - Are able to separate your personal opinion from the assigned persona and evaluation criteria Requirements - Have advanced or fluent English - Be comfortable discussing political topics in a structured and professional way - Be able to follow detailed project guidelines precisely - Be able to complete realistic role-play based on assigned political scenarios - Understand that using AI during the application process or during the project is strictly forbidden Important Compliance Requirement The use of AI tools is strictly forbidden in both the application process and the project tasks. This includes using AI to write application answers, translate messages, create conversation turns, draft evaluations, or produce justifications. All work must be written by the contributor. If AI use is detected, the contributor may be removed from the selection process or from the project. Compensation - Brazil: R$ 100 per approved task - Other countries: $20 per approved task Only approved tasks are eligible for payment. To increase the chances of approval, contributors must carefully follow the guidelines. Final Note This is a high-impact AI safety project focused on political neutrality. Your work will help evaluate how AI systems respond to users across different political perspectives, especially in realistic, multi-turn conversations where disagreement, persuasion, and pushback may occur.

Role Description We are recruiting legal professionals to review and improve real-world legal scenarios used to train AI assistants. Your job is not to write textbook definitions. It is to think like a senior legal practitioner: - Identify flawed reasoning - Jurisdictional inaccuracies - Procedural gaps - Ethical issues - Upgrade weak scenarios to truly test legal judgment We have multiple demands from multiple countries, so we need reviewers who are comfortable working with data and context exclusively from legal scenarios across different jurisdictions. We are especially looking for professionals with strong knowledge of legal systems in the United States, Europe, Brazil, and Latin America — these are the regions where our demand is highest. Candidates who are familiar with more than one of these legal systems are highly encouraged to apply. Qualifications - Lawyers, Attorneys & Solicitors (any jurisdiction) - Paralegals & Legal Assistants - Legal Researchers & Academics - Compliance & Regulatory Specialists - Notaries & Judicial Officers - Anyone who works with legal documents, contracts, or case analysis professionally Requirements - If selected, you will be invited to review real legal case scenarios as part of the project. Benefits - Payment will be per approved task, converted and paid in your local currency. - The correspondent payment will be approximately US$ 50 per hour. - If you would like to refer someone, you can earn $20 USD for each approved referral. Company Description Vetto is a global platform that connects top-tier professionals to strategic Artificial Intelligence projects around the world. Our mission is to build trust, quality, and long-term value within the AI ecosystem, for both exceptional talent and companies operating at the forefront of technology.

Role Description We're recruiting engineering experts to review and improve real-world technical scenarios used to train AI planning assistants in an educational context. The AI model will act as a tutor — explaining engineering topics, solving problems step by step, and teaching concepts to university-level students. Your job is to think like a senior engineer and educator: - Map decision trees - Identify alternative hypotheses - Justify conclusions with concrete data - Ensure reasoning is technically sound and clear enough to teach Qualifications - Mechanical, Civil, Electrical, Chemical, Industrial, or Aerospace Engineers - Any engineer with hands-on experience diagnosing problems, designing solutions, or running technical analyses - Professionals who routinely make data-driven decisions in their field - Final-year undergraduate engineering students with a solid technical foundation are also welcome to apply Requirements - Present a real technical engineering problem you diagnosed or solved — a failure, root cause analysis, dimensioning decision, process issue, etc. - Describe the technical problem and what data or information you had available at the start. - Explain your reasoning in 3 steps, detailing what you analyzed or decided and what specific measurement, calculation, test, or evidence drove that decision. - For each of the 3 steps, list at least 2 hypotheses you considered but ruled out and explain what concrete data eliminated each one. - Describe the final diagnosis or solution and how the evidence you gathered led to it. Benefits - Payment will be US$ 60 per approved task, converted and paid in your local currency. - Each task takes approximately 80 minutes, which corresponds to an effective rate of about US$ 45 per hour. - If you'd like to refer someone, you can earn $20 USD for each approved referral.

Role Description The goal is to create gold-standard answer keys and grading rubrics for intentionally ambiguous finance questions, grounded in real company filings. - Define and justify appropriate and essential metrics and KPIs by sector / stage / business model - Interpret 3–4 years of results, including one-offs, accounting effects, mix, and cycle - Apply sector-specific repertoire: drivers, essential KPIs for that market, red flags, and typical patterns Qualifications - Minimum of 4 years of experience in the field - Strong command of corporate finance and market analysis - Above-average depth in at least one GICS sector - Advanced English Requirements - This role is designed for senior-level experts with real market experience Benefits - Remote and asynchronous work model - Estimated dedication: ~10 hours total (may extend up to ~15h depending on pipeline needs) - Work format: task-based (typical task: 30–60 minutes) - Compensation: USD 100/hour

Role Description This role sits at the heart of Vetto’s mission: using high-quality human data to build AI systems that make the world better. You’ll take raw expert signals and turn it into tangible model improvement, experimenting rapidly and carving new paths in post-training. With full autonomy and no production constraints, you’ll have the freedom to try unconventional ideas and see their impact quickly. Key Responsibilities - Design and run post-training experiments on frontier and open-weight LLMs (SFT, preference-based methods, rubric-driven training) - Translate raw annotation artifacts (multi-step solutions, evaluations, adversarial prompts) into training-ready datasets - Prototype new reward signals beyond pairwise preferences (rubrics, constraints, structured critics) - Analyze failure modes; propose data-centric fixes (sampling, curriculum, counterfactuals) - Build lightweight training/eval pipelines; iterate quickly - Produce short internal memos: what worked, what didn’t, why Qualifications - PhD (or equivalent experience) in ML/AI, applied math, stats, or adjacent - Hands-on experience with LLM post-training (at least one of SFT/DPO/RLHF/RLVR) - Solid Python + PyTorch/JAX; comfortable with training infra basics - Fluent English Preferred Qualification - Worked with rubric-based evaluation or tool-augmented tasks - Experience mixing synthetic and human data - Familiarity with failure analysis and dataset audits Work Model We operate remote-first. We focus on outcomes, not where the work is done. To support flexibility and personal choice, we maintain offices in select locations as an optional resource for the team. Location Flexible (EU-friendly time zones preferred) Type Full-time or long-term contract Equal Employment Opportunity Vetto is proud to be an equal opportunity employer and values diversity at our company. We do not discriminate on the basis of race, color, religion, national origin, sex, sexual orientation, gender identity, age, disability, veteran status, or any other protected characteristic.

Role Description We’re recruiting QA (Quality Assurance) professionals to support multiple projects at Vetto. In this role, you’ll: - Review and audit responses and tasks produced by human annotators. - Ensure the final delivered data meets our highest quality standards. - Help identify content that may have been generated by AI or that falls outside established guidelines, standards, and evaluation criteria. Qualifications - Experience in Quality Assurance or a related field. - Strong attention to detail. - Ability to work independently and as part of a team. Requirements - All answers MUST be in English. - Do not use AI; own the task. - Respond calmly and in detail to application questions. Benefits - Earn $20 USD for each approved referral.

Role Description We're recruiting management consulting experts to review and improve real-world business scenarios used to train AI planning assistants in an educational context. The AI model will act as a tutor — structuring complex problems, building hypothesis trees, and teaching strategic reasoning to business students. Your job is to think like a consultant at a top-tier firm: - Break down ambiguous problems - Structure your approach - Identify and discard hypotheses with data - Deliver a clear recommendation Qualifications - Current or former consultants from strategy and management consulting firms (e.g. McKinsey, BCG, Bain, Roland Berger, Oliver Wyman, Accenture Strategy, or similar) - Professionals with experience in structured problem solving, issue trees, and hypothesis-driven analysis - MBA students or final-year undergraduates with strong case interview or strategy project experience Requirements - Present a real business problem you diagnosed or solved — a strategic decision, market entry, operational inefficiency, organizational challenge, etc. - Structure your case in 4 parts: - Part 1 — The Problem: Describe the business problem and what data or information you had available at the start. - Part 2 — Your Journey: Describe your reasoning in 3 steps. For each step, explain what you analyzed or decided and what specific evidence, data, or framework drove that decision. - Part 3 — Discarded Alternatives: For each of the 3 steps, list at least 2 hypotheses you considered but ruled out and explain what concrete data eliminated each one. - Part 4 — Conclusion: Describe the final recommendation or solution and how the evidence you gathered led to it. Highlight 1–2 key insights. Benefits - Payment will be US$ 60 per approved task, converted and paid in your local currency - Each task takes approximately 80 minutes, which corresponds to an effective rate of about US$ 45 per hour