DEV Community: Vani Chitkara

Ending the 2 AM Nightmare: How My Backtrace Agent and GitLab Orbit Tame On-Call Chaos

Vani Chitkara — Sun, 21 Jun 2026 19:34:37 +0000

We have all been there. It is 2:00 AM, and your phone starts screaming on the nightstand. You squint at the screen to see a cryptic alert: "login error rate jumped 5x." Your heart sinks because you know exactly what comes next. You crawl out of bed, open your laptop, and begin the frantic detective work.

You start digging through deployment logs and scrolling through a long list of recent Merge Requests. You check Slack to see who was active late in the day. You are under maximum pressure to figure out what shipped, which change touched the login flow, and who wrote the code. It is an hour of stress and guesswork while the error rates continue to climb.

This is the "on-call tax" that every developer pays, but it does not have to be this way. To solve this, I created Backtrace, a specialized GitLab agent and flow designed to turn that hour of panic into seconds of clarity.

What is Backtrace?

Backtrace is a GitLab Duo Agent and an automated flow that does the heavy lifting for you. Instead of leaving you with a wall of recent deploys and wishing you good luck, it traces a production incident backward through the software lifecycle.

When a production incident is opened, the Backtrace flow automatically assembles the answer. For example, in a real scenario where login failures are spiking, Backtrace can look at the data and tell you exactly what happened. It might report that login failures started right after the latest deployment to production. It identifies that the deploy shipped a specific Merge Request, which changed a file called auth/session.py on the failing path. It even tells you the author and the specific work item, such as "Tighten session expiry," so you know exactly where to start.

The Secret Sauce: GitLab Orbit

You might wonder how Backtrace is different from other AI tools. Most tools rely on LLM guesswork or simple keyword matching. Backtrace is different because GitLab Orbit powers it.

GitLab Orbit is a queryable knowledge graph that maps the hard facts of your development process. It connects the dots between environments, deployments, merge requests, changed code, and the people who wrote it. Without Orbit, this level of automation simply could not exist because no other tool maps deployments to code changes in one unified graph. My Backtrace agent uses these verifiable graph facts to follow every hop from the production crash back to the specific line of code that caused it.

Beyond Just Finding the Problem

Finding the problem is only half the battle. When production is broken, every second counts. Backtrace does more than just point a finger. It takes four instant actions to help you fix things:

Traces the Graph: It maps the entire path from the environment failure back to the user who wrote the code.
Ranks the Culprit: It matches the incident symptoms, like a login spike, against the files that were recently changed to find the most likely cause.
Names a Rollback Target: It identifies the last good deployment that was running before things went wrong, so you know exactly where to revert.
Pages the Right Humans: It assigns the incident to the original author, mentions the person who deployed it, and applies triage labels.

The Bottom Line

By leveraging the power of GitLab Orbit, the Backtrace agent and flow, change the narrative of incident response. It moves us away from frantic searching and brings us towards fact-based mitigation. When that pager goes off at 2 AM, you won’t be starting a search from scratch. Instead, you will be looking at a solution that is already prepared for you.

Learn more about Backtrace: https://youtu.be/QUAIEHj3mVw

Building Parallax: The Vision-Powered UI Navigator Agent

Vani Chitkara — Sun, 15 Mar 2026 16:08:18 +0000

This piece of content was created specifically for the purposes of entering the Gemini Live Agent Challenge hackathon. #GeminiLiveAgentChallenge

Traditional automated testing is broken. It relies on "cheating" by looking at the underlying HTML code (the DOM) to find buttons and links. But humans don’t browse the web by reading code; we browse by seeing pixels.

When we (my teammate and I) set out to build Parallax, we wanted to create a truly human-centric testing agent. We didn't want a scraper; we wanted an agent with "eyes." To achieve this, we turned to the cutting-edge capabilities of Google Gemini 2.5 Flash and the Google Cloud ecosystem.

🧠 The Core: A Vision-to-Action Brain

At the heart of Parallax is the Gemini 2.5 Flash model. We chose this model specifically for its industry-leading multimodal performance and low latency.

In Parallax, we don't send a single line of HTML to the AI. Instead, our agent loop performs the following:

Capture: Using Playwright, we snap a high-resolution screenshot of the browser viewport.
Analyze: We send that raw image to gemini-2.5-flash with a specific user persona context (e.g., "You are Martha, a 72-year-old with low tech literacy").
Act: The model "sees" the UI elements and returns raw pixel coordinates for the next action—be it a click, a scroll, or a type. By using gemini-2.5-flash, the agent can identify UX friction that code-based tests ignore, such as poor color contrast, overlapping elements, or confusing visual hierarchies.

🛠️ Multi-Agent Orchestration with Google ADK

Parallax doesn't just run one test; it runs a "swarm" of diverse perspectives. We used the Google Agent Development Kit (ADK) to orchestrate these independent persona agents. The ADK allowed us to create distinct cognitive models for each persona, ensuring that "Martha" (our 72-year-old dear grandmother), "Raj" (our 28-year-old power user), and our 5 other agents with diverse personas can navigate the same site simultaneously, each reporting unique findings based on their specific technological background.

📈 Scaling on Google Cloud

To handle the intensive compute requirements of headless browsers and high-frequency AI calls, we built a serverless architecture on Google Cloud:

Google Cloud Run: Our FastAPI backend is fully containerized and deployed on Cloud Run. This allows us to scale horizontally as more agents are spawned, ensuring that the "Vision Loop" remains snappy and responsive.
Google Cloud Firestore: We use Firestore for real-time state management. As agents find issues, they are instantly streamed to a live dashboard, allowing developers to watch the "thinking process" of the AI in real-time.
Google Cloud Storage (GCS): Every multimodal artifact—every screenshot the agent "saw" is persisted in GCS. This creates a visual audit trail that is invaluable for debugging UX failures.

💡 Conclusion

Parallax represents a shift from "testing code" to "testing experiences." By combining the multimodal power of Gemini 2.5 Flash with the reliability of Google Cloud, we’ve built a tool that helps developers see their apps through the eyes of their most diverse users.

Check out the live project here: https://bit.ly/parallax-agent

Building PersonaPrep: An AI Personality Coach with Kiro

Vani Chitkara — Sat, 06 Sep 2025 10:02:14 +0000

In a world where social cues, confidence, and culture shape our day-to-day interactions, one thing remains clear: effective communication is a superpower.
Yet, for many people, initiating conversations, navigating interviews, or simply showing up confidently in new environments—like a first day at school, college, or job—can feel like climbing a mountain.

🧠 What is PersonaPrep?

PersonaPrep is an AI social coach that adapts to your personality type and helps you practice high-stakes or awkward social scenarios through interactive, real-time simulations.

Whether you're:

Interviewing for your dream job,
Speaking up in a meeting,
Making friends in a new country,
Or overcoming social anxiety,

PersonaPrep lets you rehearse, refine, and reflect—all in a judgment-free, deeply personalized way.

🧩 Why an AI Personality Coach?

PersonaPrep focuses on three big problems:

Social anxiety → Rehearsing workplace introductions, meetings, or even casual small talk.
Adaptive learning → Tailoring practice sessions based on user needs and past conversations.
Language barriers → Practicing common phrases in a new language or cultural context.

Instead of a static chatbot, PersonaPrep is like a personal coach that learns and guides you to drive conversations confidently in a new environment.

⚙️ The Tech Stack: Fast, Flexible & Real-Time

Layer	Stack
Backend	Spring Boot (Java 17) with REST APIs + WebSocket support
Frontend	React + Tailwind for snappy, intuitive UI
Database	MongoDB Atlas to persist sessions and coaching history
LLM	Gemini 2.0 Flash for AI-powered coaching

📊 How It Works: Flow from Personality to Mastery

PersonaPrep follows a step-by-step, coach-guided journey:

1️⃣ Practice Conversations

Simulate real conversations with diverse AI characters:

Interviewers
Classmates
Managers
Friends or strangers

AI agents are steered by personality context, dynamically adapting tone, topic depth, and challenge level.

2️⃣ Feedback & Growth

We log:

Confidence scores
Emotional intelligence traits
Communication effectiveness

Like Kiro’s auto-generated spec summaries, this feedback is contextualized and actionable—offering insights like “Try pausing before responding” or “This phrase could sound more confident.”

3️⃣ Mastery & Beyond
As users grow, they unlock:

Advanced role-plays
Peer coaching
Opportunities to help others—becoming mentors in their own right

🧬 How Kiro Inspired PersonaPrep’s Architecture

Kiro’s philosophy of spec-first, agent-powered development had a profound influence on our build.

✅ Spec-Driven Conversations

Just like you define a spec in Kiro before coding, our system defines interaction goals (e.g., “Nail a behavioral interview answer”) before launching a coaching session. This “spec” feeds into AI steering logic to ensure alignment.

🔁 Agent Hooks & Triggers

We use Kiro-style hooks to automate:

Feedback generation after a session
Session summarization
This enables an agentic feedback loop—users practice, reflect, and try again, with minimal friction.

🧭 Steering Personalities

Kiro uses project context and conventions to steer agents. We use personality styles (outgoing, introverted, anxious, task-oriented) to steer our coaches. This ensures communication feels natural and aligned to the user.

For example:

An anxious user might start with low-pressure scenarios.

A tool-oriented user gets measurable milestones and structure.

Benefit	Result
Backend	Practice interviews, presentations, and meetings
Social Comfort	Engage in small talk, new cultures, or dating
Emotional Intelligence	Understand how you sound, learn empathy
Personal Growth	Track progress, unlock badges, and celebrate milestones

📖 Lessons Learned

Kiro as a teammate: Beyond just code, Kiro gave us structure, specs, and clarity.
AI accelerates hackathons: With boilerplate handled, we spent time on features that mattered.
Scope discipline matters: Steering docs ensured we built something shippable, not just “cool demos.”

💬 Final Thoughts

PersonaPrep wasn’t just a hackathon project — it showed us the power of AI tools like Kiro to bridge the gap between ideas and production-ready code.

With Kiro’s spec mode and steering docs, we shipped a full-stack AI application in record time.

If you’re building at hackathons (or even production projects), Kiro isn’t just an AI assistant — it’s a project accelerator.