A year of using ChatGPT as your second brain for deals, frameworks, and family life. From AI-native VC theses and secondaries to Finland cabins, telescopes, and health experiments—this is how you actually used your copilot in 2025.
You rarely ask for a one-off answer. You ask for reusable systems—ontologies, frameworks, checklists—that you can keep reusing across work and life.
You leaned on ChatGPT as a high-leverage thinking partner:
You rarely copy answers blindly; you remix them into your own systems, emails, and frameworks.
You think in systems, not one-off answers
You used ChatGPT to sketch a global VC ontology, JSON-LD schemas, and AI-enabled workflows for sourcing, memos, and decision rationales. The goal wasn’t just ‘better notes’, but a reusable investment OS that compounds over time.
Thinking in scenarios, carries, and optionality
You explored secondaries (e.g. AI infra players), partial take-profit frameworks, and structures like carry taxation and BVI/US/Lux vehicles. You repeatedly asked for matrices and factor models, showing a strong risk/return and scenario mindset.
Sharp but kind communication
From OpenAI allocation updates to warm intros between founders and investors, and positioning EGV in a sentence, you used ChatGPT to tune language across EN, PT-BR, and IT. Tone: respectful, concise, human, no fluff.
Optimizing for memories per euro, not just logistics
You designed trips where constraints really matter: kids at different stages, an elderly mother, seasons, budgets, and energy levels. Think Hungary, Finland cabins, Bali vs Thailand, Prague school trips. You asked for itineraries that mix culture, nature, and realistic pacing.
ROI mindset in the physical world
From dishwashers and telescopes to property and renovation ROI, you consistently framed home decisions as small capital allocation problems: price/value, durability, energy class, and long-term convenience.
From heart rate formulas to thyroid nuance
You went beyond gadgets: HR formulas (Tanaka, Karvonen), heart rate reserve, hypothyroidism (Bianco), supplements, magnesium timing, running at -10 °C, and devices like Hume / Whoop / Garmin. The pattern: evidence-aware curiosity and long-term health horizon.
Your default question: ‘How do we generalize this?’
You often requested matrices, stepwise frameworks, JSON-LD ontologies, templates, decision trees, and scoring models. It’s less ‘What should I do?’ and more ‘How should I reason about this class of problems forever?’
PT-BR, IT, EN as a single workspace
You flowed between Portuguese, Italian, and English depending on audience: LPs, co-investors, school parents, lawyers, and founders. ChatGPT became a translation layer for your cross-border life and work.
High bar for clarity, zero tolerance for coldness
You repeatedly refined emails and messages until they felt simultaneously clear, concise, and genuinely warm. This is a consistent signature: you care that the other side feels respected and not overwhelmed.
If you keep following your current curiosity, your 2026 focus might naturally cluster like this:
Net-net: more leverage from systems, not more hours.
From ontology slides to an actual cockpit
2025 showed a clear pull: you want a live deal room powered by your ontology, scoring models, and RAG. In 2026, a realistic bet is shipping a minimum viable ‘Martino Deal OS’ that your team actually uses.
Standard operating procedures for memories
You’ve basically designed a template engine for trips: constraints, budgets, mobility, local culture, kid/elder-friendliness. Turning this into checklists and templates would reduce friction every time you plan something new.
Less mental overhead, more consistent data
You already track HR, thyroid nuance, supplements, and wearable data. A simple, visual dashboard—built once, reused forever—would reduce cognitive load and make patterns obvious.
Glue your tools together
You’ve explored RAG, Canvas, ontologies, and integrations. A 2026 experiment is a thin glue layer that routinely syncs deals, notes, and contacts into your VC brain without manual friction.
Light-hearted, but pretty plausible
You’ll ask if a destination is both kid- and grandma-friendly at least three times. You’ll iterate your VC ontology again, closer to a product. And you’ll keep using ChatGPT for more contracts, structures, and ‘translation’ between worlds.
Start here: a snapshot of how you and ChatGPT worked together in 2025.