Principal Software Engineer

Gitwit is an AI-first venture studio that finds problems worth solving, rigorously validates them, and builds AI-native companies. The Principal Software Engineer will shape products and ventures from zero to one, set technical direction, build and ship full-stack AI applications, and collaborate across product, design, research, and venture leadership.

Responsibilities:

• Shaping products from the beginning. Helping define what should be built, what should wait, and what should be cut before time gets wasted
• Building from zero and shipping fast. Taking ideas from early concept to prototype to real product in users’ hands, often in days, not months
• Seeing your work in the wild immediately. You won’t work on tiny pieces of a giant system that may never reach production. What you build here ships, gets used, and teaches us something right away
• Making foundational technical decisions. Choosing the right tools, architecture, and tradeoffs for the stage, instead of inheriting someone else’s stack by default
• Working close to the truth. Learning directly from users, pressure-testing assumptions, and adjusting quickly when the signal changes
• Building across the full product reality. Thinking not just about whether something works technically, but whether it is useful, usable, differentiated, and economically sound
• Operating at a pace that keeps things fun. Small teams, minimal process, and the expectation that meaningful progress happens daily. If you love moving quickly, you’ll thrive here. If you prefer long cycles and layers of review, this will feel uncomfortable
• Helping shape ventures, not just software. Contributing to the decisions that determine whether something becomes a feature, a product, or a company

Requirements:

• This role is designed for engineers who have already led major technical initiatives, shaped engineering strategy, and demonstrated exceptional judgment in ambiguous zero-to-one environments
• We are looking for someone who has personally done meaningful zero-to-one work
• Someone who has helped shape the product, technical, and execution decisions that determine whether something gains traction or stalls
• That might mean you were an early engineer at a startup
• It might mean you built a new product or system from scratch inside a larger company
• It might mean you created something on your own and carried it far enough to learn what real users, real constraints, and real tradeoffs actually feel like
• What matters is that you have felt the difference between building from zero and executing inside something already defined, and you know you want more of the former
• They have built something from scratch far enough to make meaningful product, technical, and speed-versus-quality tradeoffs in the real world
• They care about what should be built, what users actually need, what should be simplified, and what is not worth building
• They do not need every variable resolved before they can make progress. They know how to create clarity while building
• They can make sound architectural and tooling decisions for the stage, without over-engineering or blindly defaulting to whatever is newest
• They have experience designing and shipping full-stack applications using modern frameworks and understand how to balance speed, maintainability, and product quality
• They have experience working with foundation model APIs and understand concepts like prompt engineering, retrieval-augmented generation (RAG), tool calling, agent workflows, evaluation, and the practical tradeoffs involved in bringing AI features into production
• They stay close to the frontier of new AI capabilities, tools, interfaces, and product patterns, not to chase novelty, but to understand what creates real leverage. They know how to evaluate new technology quickly, apply it selectively, and use it where it makes the product sharper, the workflow stronger, or the venture more advantaged
• They care whether something is sharp, usable, coherent, and worth a user's time, not just whether it technically functions
• They communicate clearly, collaborate well, and can think across product, user, business, and technical realities without losing rigor
• If you thrive in high-autonomy environments, communicate clearly, and can drive zero-to-one work without needing constant supervision, this setup is designed for you
• They have real zero-to-one evidence
• They think like a product builder, not just an implementer
• They move well in ambiguity
• They have strong technical judgment
• They have built modern web products
• They have built AI-native product experiences
• They are voracious learners with sharp judgment about emerging technology
• They have taste
• They can work as part of a real venture team
• They want the real version of ownership