Machine learning (ML) is the field of computer science focused on building systems that learn patterns from data rather than following explicitly programmed rules. It encompasses a broad spectrum from classical statistical methods (linear regression, decision trees, clustering, random forests) through deep learning (neural networks with many layers) to the modern generative AI and foundation models that dominate current attention. ML is the parent field within which generative AI is a recent and currently dominant subcategory. Most ML work today is still classical or traditional deep learning; generative AI is the most visible but not the totality of ML.
The main categories of machine learning:
Supervised learning: train on...
Valuation is the estimated worth of a company at a specific moment, calculated through multiple methods. Methods include comparables analysis (looking at recent transactions of similar companies), discounted cash flow (modeling future cash flows back to present value), market-based pricing (what the marginal investor will pay), asset-based methods (for asset-heavy businesses), and for early-stage startups, venture-style heuristics anchored on round dynamics, growth-stage benchmarks, and qualitative assessments of team, market, and momentum. It is the number that shapes every fundraising conversation, every employee equity grant, every acquisition discussion, and every founder anxiety about whether they're being underpriced or over...
Product discovery is the practice of validating problems, opportunities, and solutions with customers and data before committing engineering effort to build them. It is distinguished from product delivery (the act of designing, building, and shipping) and aimed at killing bad ideas cheaply so the team only builds things likely to drive the intended outcome. It is the front half of modern product work, and the part most under-invested in by startups that mistake speed of shipping for speed of learning.
The discipline was popularized in its modern form by Marty Cagan, especially in Inspired (2008/2017) and Empowered (2020), with the core argument that great product teams discover what to build before they decide to build it,...
An Employee Stock Purchase Plan (ESPP) is the broad-based benefit that lets employees purchase company stock at a discount via payroll deductions. Typically established under IRC Section 423 for qualified tax-favored treatment, employees buy stock at up to 15% below market price at the end of each 6-month offering period, almost exclusively at public companies because the mechanic requires liquid stock. It is a meaningful employee benefit at public companies and largely irrelevant at private startups before IPO.
The Section 423 (qualified) ESPP mechanic:
Cap table cleanup is the pre-financing process of resolving stockholder ambiguities, missing signatures, expired options, and inconsistencies before investor diligence reveals them. The cleanup covers improperly granted options, undocumented promises of equity, lost stock certificates, and other inconsistencies in the company's capitalization records, typically done in the 4-8 weeks leading up to a priced round and one of the most-overlooked but most-impactful pre-financing activities. It is the difference between a smooth diligence process that closes on schedule and a contentious one where issues surface at the worst possible moment.
The common issues that surface in cap-table cleanup:
Wefunder is the largest equity crowdfunding platform operating under Regulation Crowdfunding (Reg CF) by total raise volume. It has facilitated approximately $700 million in capital across thousands of startup offerings since founding in 2012 (predating the formal Reg CF rollout in 2016), with a portfolio that has included many Y Combinator alumni, B Corp companies, mission-aligned ventures, and consumer brands with passionate customer bases. Wefunder is itself structured as a public benefit corporation, reflecting the platform's emphasis on democratizing startup investment access.
The structural characteristics: Reg CF focus (most offerings are Reg CF; Wefunder also supports Reg A+ for larger raises). Standard offering size typica...
An Incentive Stock Option (ISO) is the tax-advantaged stock option granted only to employees under IRC Section 422. It produces no ordinary-income tax at exercise (subject to AMT) and qualifies for long-term capital-gains treatment on the entire spread if held two years from grant and one year from exercise, capped at $100,000 of FMV first-exercisable per employee per year. It is the more tax-favored of the two main option types and the default for employee grants at most C-corp startups.
The ISO mechanic and the tax rules:
Annual planning is the yearly planning cycle that sets direction, goals, resource allocation, and budgets for the year ahead, typically run in Q4. It brings together strategic vision, financial planning, hiring plan, product roadmap, and team OKRs into a coherent annual operating plan. Annual planning becomes load-bearing at growth-stage companies (50+ employees) and is generally too formal at very early-stage, where strategy needs to iterate faster than annual cycles. It is the planning anchor for growth-stage companies and the document that ties strategy to execution.
The typical annual planning process:
Phase 1: review and reflection (4-6 weeks before fiscal-year start):
User research is the practice of generating evidence about user needs, behaviors, and pain points through systematic methods, then synthesizing that evidence into actionable insights. Methods include interviews, observation, surveys, usability testing, diary studies, and analytics, and the output informs product, design, and marketing decisions. It is the input layer underneath product discovery, UX work, value proposition development, and most credible go-to-market positioning.
The discipline splits along two axes that matter. The qualitative versus quantitative axis: qualitative research (interviews, observation, usability tests) tells you why users do what they do and is best run in small N with rich detail; quantitative re...
Issued shares is the total number of shares a corporation has actually issued to stockholders across the company's history. It is distinct from authorized shares (the legal maximum permitted by the charter) and outstanding shares (issued minus treasury shares), counted cumulatively before any reductions for repurchased or canceled shares. It is one of three closely-related but distinct share-count concepts that founders need to understand for accurate cap table management.
The three share-count concepts: