Vamsi Narla

I asked ChatGPT a question millions of job seekers now ask it: what’s the best AI tool to practice for interviews? It gave me a confident, well-formatted list. Then I did the thing the chatbot never does — I opened every link. Otavo, Zilta, Prepra, VocalHyre — polished landing pages, the right keywords, “AI mock interview, real-time feedback, STAR method.” No company behind them. No founder. No reviews anywhere a stranger left them. Zilta even turned up on a marketplace for selling unfinished side projects. Yoodli, the tool it recommended most confidently, had quietly pivoted to enterprise sales coaching; interviews are now one line on its homepage. Google’s Interview Warmup — the “best free option” — was shut down months ago. The old URL redirects to an article. ...

The hardest engineering problem in building Revarta wasn’t making the AI smart enough to evaluate an interview answer. The models are already good at that. It was making the AI willing to tell someone their answer was bad. Ask a stock model to grade an interview answer and it will tell you almost anything is “a strong response.” A vague, rambling story with no result lands a cheerful 4.5 out of 5. This isn’t a capability gap — the same model can dissect exactly why the answer is weak if you ask it to critique a stranger’s. It’s a disposition. These models are trained, through human feedback, to be agreeable, and that training has gravity. It’s the same instinct that makes a coding agent say “you’re absolutely right” the moment you push back on its design — it would rather agree than hold a line. ...

I’ve interviewed more than a thousand people. I can’t tell you the middle of almost any answer I heard. What I can tell you, often months later, is how a candidate opened and how they landed. That’s not a flaw in my memory. It’s how memory works. We don’t store experiences as continuous recordings — we keep the start, we keep the end, and we compress everything in between. Kahneman called the end-part of this the peak-end rule; the front-loaded half is just as real. First impressions are a cliché because they’re true. The part most people miss is the symmetry: the last impression matters every bit as much, and the messy middle barely registers. ...

A product can be perfectly compliant — every screen reader satisfied, every contrast ratio passing, every guideline met — and still be completely out of reach for the person who needs it. When most people hear “accessibility,” they think of the first part: designing for people with disabilities. That work is real, it matters, and I’m not here to relitigate it. There’s a second sense of access that gets far less attention: whether a genuinely good thing is reachable by an ordinary person at all. Not “is the interface usable,” but “can you get this in the first place.” By that measure, most high-quality services quietly fail. The best interview coaching runs two hundred dollars an hour and books weeks out. The best of almost anything is gated — by price, by availability, by the unspoken assumption that excellence is for people who can pay for it. ...

The interviewer asks “tell me about a time you led through conflict,” and your mind goes blank. Not because you’ve never done it. Because you tried to store fifteen years of experience in the worst possible place — your memory, retrieved live, under pressure. Most interview prep is top-down. Memorize STAR. Memorize formats. Memorize what a good answer sounds like. That’s optimizing the wrong layer. The candidates who walk in calm did the opposite, bottom-up work: they built a catalog of their own stories before they needed it. ...

Working with agentskills.io and skills.sh reminds me of the “Not yet” line in The Matrix. Trinity is on a rooftop, in the cockpit of a helicopter she’s never seen. Neo asks if she can fly it. She says “Not yet” and gets a download of a pilot program (skill). A few seconds later she is airborne. I bring this up because recently as I code, I’ve been religiously adding skills to the code base and somehow that act of finding the right skill for work I’m doing — working with gemini live API, composio.dev or supabase postgres tables — reminds me of this scene from the Matrix. ...

A few months back a friend shared Steve Yegge’s Welcome to Gas Town post with our group and asked where everyone is. The post lays out a ladder for the AI coding process - eight stages, from “barely using it” to “running your own orchestrator.” Stage 1: Zero or near-zero AI Stage 2: Coding agent in the IDE, permissions on Stage 3: Same agent, YOLO mode Stage 4: Wide agent in the IDE - code is mostly diffs Stage 5: CLI, single agent, YOLO Stage 6: CLI, multi-agent, YOLO - three to five in parallel Stage 7: Ten-plus agents, hand-managed Stage 8: Building your own orchestrator When my friend shared the original post, I was transitioning to Stage 6/7 so this post is a bit delayed because I wanted to share the state only when I’m comfortable where I’m at. For me each step was a process, way of working and ultimately a mindset leap. ...

When an AI product produces good output, three things had to go right. When it produces bad output, the cause is almost always in one of those three things. The model itself, Claude or GPT or Gemini or whichever, is the most visible variable. It’s rarely the most consequential. ...

A few years ago I developed EFQ, a prioritization framework for deciding what to build. It stands for Effort, Frequency, Quality risk. I’ve been applying it to automation and agentic builds. It translates directly. What is EFQ? EFQ is modeled on the ICE scoring approach where you score ideas to quantitatively determine which to pursue. The difference: all three factors come from the customer’s perspective, not yours. Effort — How much time does this task take the customer? Minutes = 1, hours = 3, a full day = 5. ...

As a leader of a portfolio of products and areas, director or VP, the breath and depth of what you lead and with the team size you lead it is important to have the right structure in place to support the team(s) and the business. While right team structures are not a guarantee to success getting this part wrong can definitely create obstacles towards success. Recently adopted this model framework and thought it’s worth sharing. ...