First principles thinking—breaking down problems to their fundamentals rather than relying on past experience—is emerging as a critical skill for engineers navigating rapid technological change, according to a post by Sunil Sadasivan.
Sadasivan reflects on what distinguished the best senior engineers he’s worked with, noting they shared a common habit: “they seemed to know what needed to be done.” Rather than this being innate talent, he traces it to a deliberate approach of asking why something is being built and what it does for users, then connecting codebase decisions to real-world outcomes. “That understanding helped them keep things simple,” he writes.
This approach builds momentum through incremental progress. Instead of getting stuck trying to solve everything at once, breaking work into the smallest accomplishable task helps engineers figure out next steps. The pattern applies whether dealing with established constraints or novel problems.
The transition to agentic AI development makes this skill particularly urgent. Sadasivan describes it as a “simulation switch-up” requiring engineers to temporarily set aside accumulated knowledge. “The engineers I see keeping up with what’s possible are willing to put what they know in a box for a while as they work with agents,” he notes, adding they’ll test assumptions rather than assuming old constraints still apply.
He identifies a core challenge: past experience and technical limitations can answer a question before the problem has been fully understood. “When I step back and ask what we’re actually trying to do, why it matters, and how the pieces connect, I usually find more ways forward than I expected,” Sadasivan writes.
Applying first principles thinking to AI involves starting with goals rather than excitement about the technology. This approach naturally aligns with the momentum-building methodology—once you understand what you’re trying to accomplish, taking small steps, learning, and iterating becomes natural. Sadasivan argues that agentic tools enable faster learning loops when grounded in this deeper understanding, creating what he describes as “the new flow state.” The skill appears less about raw technical knowledge and more about maintaining intellectual flexibility as tools and constraints evolve.
Key facts
- First principles thinking involves understanding why something is being built and what it does for users, not relying solely on past experience
- The best senior engineers Sadasivan worked with shared the habit of thinking from fundamentals and keeping solutions simple
- In agentic AI development, engineers keeping pace are willing to temporarily set aside accumulated knowledge to explore new possibilities
- First principles thinking enables faster learning loops and momentum when working with agentic tools