Bogomil Balkansky is a Bulgarian-born Sequoia Capital partner who describes himself as a "feeler" that makes investment decisions based on intuition rather than spreadsheet analysis, which is uncommon among venture investors. He brings unusually broad enterprise operating experience from McKinsey through senior product roles at VMware, Google, and startups, combined with a distinctive focus on helping founders "iterate quickly and sequence your way to the big vision" rather than trying to build everything at once.

• A strong emphasis on re-engineering the software development lifecycle to be AI-native rather than just adding tools. He frequently highlights that integrating artificial intelligence requires a fundamental process shift, such as requiring detailed documentation and context before coding even begins. He points to Salt Security as a prime example of this, noting their strict internal rule where engineers cannot demo a feature unless the AI-generated requirements and design documents are created first.
• Focus on the evolving role of the software engineer from code writer to architect and reviewer. He argues that while AI handles the heavy lifting of writing code, the human engineer must retain total ownership of quality and reliability. He illustrates this with a metaphor comparing AI to a race car, suggesting that in the hands of a skilled driver it provides a massive speed boost, but without the right guardrails and skills, it causes accidents.
• Interest in the infrastructure required to productionize AI agents and applications. He looks for the underlying plumbing that makes AI reliable in a corporate setting, drawing parallels to how GPUs became the unexpected infrastructure for the current tech boom. He specifically mentions Temporal as a company that wasn't originally built for AI but has become essential infrastructure for managing reliable AI agents.
• Belief that context engineering is more important than prompt engineering. He repeatedly advises that feeding the model the right background information, architecture diagrams, and data flows is the biggest lever for success. He details technical strategies like auto-generating markdown files for every repository to ensure the AI understands the high-level architecture before it attempts to solve a problem.
• Attention to the changing landscape of cybersecurity, specifically the concept of fighting autonomous threats with autonomous defense. He notes that as bad actors use automation to launch attacks, security teams must deploy AI agents to counter them. He frequently mentions Wiz and newer investments like Traversal AI to highlight the need for automated troubleshooting and defense mechanisms.
• A prioritization of speed and removing dependencies in go-to-market strategies. He suggests that internal friction between teams kills growth and supports platforms that allow non-technical teams to move faster without waiting on engineers. He uses Mutiny as a case study for empowering marketing teams to execute campaigns and personalization independently to avoid bottlenecks.