DEV Community: NIRANJAN LAMICHHANE

Coffee, Code, and Junie: My Journey to 10x Productivity with JetBrains AI

NIRANJAN LAMICHHANE — Tue, 03 Feb 2026 14:25:59 +0000

Dream of 10x Developer :

I've always chased that mythical 10x developer status—that magical state where code flows like water, bugs vanish before they form, and architecture emerges perfectly from the chaos.But something was holding me back to achieve this .
I'm an IntelliJ evangelist. The keyboard shortcuts are muscle memory, the UI feels like home, and Android Studio is basically family. But every time I tried AI coding assistants, I hit the same wall—they all lived in VS Code's world. The mental context switching was killing my flow state.

At this time , the intelij ide announces this hackathon of the Sky limit , providing myself the free credit and opportunity to try out junie and ai assisted coding .Its like a gift from the coding gods: free access to Junie, an AI assistant built right into my beloved IDE.

Same time , i was starting to learn python and data science too . To get maximum advantage of junie , i install python plugin in intelij ide to give a try to python and data science . Later , i found that Jetbrains have similar product “ Pycharm “ for this purposes.
Enough about me , lets talk about what I built with the help of junie .

I built ScrIM Coach, which is an esports performance analysis tool that transforms raw League of Legends match data into actionable coaching insights. We analyzes professional team performance using the GRID Esports API to identify strategic patterns, execution gaps, and draft weaknesses. We provide executive summary , detailed analysis , practice planning by analyzing the data to the coach . The coach can also print out the action plan from the insight we have constructed .

The project is licenced under MIT license in following source :

You can try out at : here

My Favourite part :

Favourite part of me was the BRAVE MODE. I give complete control over the junie to make changes to the project and its file . After providing prompt , i can have long sip of the coffee , read the hackathon details and submission requirement , while junie have completed the tasks .
I'd write a prompt like 'Refactor this to follow clean architecture with proper separation of concerns,' take a long sip of coffee, watch Junie restructure 15 files simultaneously, and return to a codebase that was suddenly... beautiful. The best part? Nothing broke. Tests still passed. The UI still rendered

Important Contribution :

The most important contribution junie make in my productivity is during the ui enhancement . Being the backend developer and android engineer previously , pixel perfect ui seems to be not my cup of tea. But using junie , i can not only draft , but also implement pixel perfect ui , in many cases , junie does it by itself , what we have to do is attach the screen shot of what went wrong and what is your desired output .

Impressive Refactoring :

The most powerful ability and impressive things about junie for me is its ability to refactor without breaking the functionality and UI . I initially designed the working MVp with some trash architecture without following any pattern and standard . I refactor the whole of my mvp to adhere to solid principle and clean code architecture in the single prompt . Interestingly , nothing breaks at all and i would be able to add new feature after that in more easier and scalable way.

UI Magic for Backend Developers:

The Flow State Achievement:

My workflow was made more smoother and i have attained the flow state during my this journey of hackathon . Adding feature , enhancing functionality were just a prompt away and its debug mode executed by itself makes it produce the code which is actually working .

Most Misused Capability :

Not only i used Junie , I also misused junie , especially in the work of version control . I ordered ( just not prompted) junie to commit the changes it has made with clear commit message and push it to the remote source . Similarly , i often ordered it to combine multiple git commit to the single one and of course , it obey me like a good servant.

Wish it was there :

I wish to make my changes live on my deployed site as soon as i make change in the codebase, because legends debug in production. If inteilij have MCP server support , it would have been few prompt away . Absence of this feature makes me little disappointed at last .

If you're like me—comfortable in JetBrains' ecosystem but curious about AI assistance—take the leap. The hackathon may be over, but the learning isn't. Junie showed me that the future of development isn't about replacing developers; it's about amplifying our humanity through technology.

My Data Science Journey: Restaurant Tips Analysis

NIRANJAN LAMICHHANE — Sun, 28 Dec 2025 08:04:57 +0000

Project: Exploratory Data Analysis on Restaurant Tips Dataset
Duration: Full EDA Process
Dataset: 243 restaurant transactions, 7 variables
Status: ✅ COMPLETED

📊 PROJECT OVERVIEW

Dataset Information

Source: Restaurant tips dataset
Initial Size: 244 rows × 7 columns
Final Size: 243 rows × 7 columns (after cleaning)
Variables:
- Numerical: total_bill, tip, size
- Categorical: sex, smoker, day, time

Project Goal

Understand what factors influence tipping behavior in restaurants through comprehensive exploratory data analysis.

🧹 PHASE 1: DATA CLEANING (Investigation 1.3)

1.1 Missing Values Investigation

Hypothesis: "The Null Hypothesis" - Why might data be missing?

What I Did:

# Checked for missing values
data.isnull().sum()
data.isnull().any()
(data.isnull().sum() / len(data)) * 100  # Percentage

Results:

✅ 0 missing values in all columns
This indicated excellent data collection quality
No imputation or removal needed

Learning Moment: Not all datasets have missing data, but always check!

1.2 Duplicate Detection

Hypothesis: Could identical transactions exist legitimately?

What I Did:

# Found duplicates
num_duplicates = data.duplicated().sum()
duplicates = data[data.duplicated(keep=False)]

# Removed them
data_clean = data.drop_duplicates()

Results:

Found 1 duplicate row
- Bill: $13.00, Tip: $2.00, Female, Smoker, Thursday, Lunch, Party of 2
- Row 198 and Row 202 were IDENTICAL
Decision: Removed as likely data entry error
Result: 244 rows → 243 rows

Key Insight: Identical transactions on same day/time are statistically improbable - likely errors.

1.3 Outlier Investigation

Hypothesis: "The Outlier Tribunal" - Are extreme values errors or legitimate?

What I Did:

# Created boxplots
plt.boxplot(data['tip'])
plt.boxplot(data['total_bill'])

# Calculated IQR boundaries
Q1 = data['tip'].quantile(0.25)
Q3 = data['tip'].quantile(0.75)
IQR = Q3 - Q1
upper_boundary = Q3 + (1.5 * IQR)

# Found outliers
outliers = data[data['tip'] > upper_boundary]

Mathematical Formula:

IQR = Q3 - Q1
Upper Boundary = Q3 + (1.5 × IQR)
Lower Boundary = Q1 - (1.5 × IQR)

For Tips:
Q1 = $2.00
Q3 = $3.56
IQR = $1.56
Upper Boundary = $5.90

Outliers Found:
| Bill | Tip | Tip % | Verdict |
|---------|--------|-------|-----------------|
| $50.81 | $10.00 | 19.7% | ✅ Legitimate |
| $48.33 | $9.00 | 18.6% | ✅ Legitimate |
| $39.42 | $7.58 | 19.2% | ✅ Legitimate |
| $48.27 | $6.73 | 13.9% | ✅ Legitimate |

Decision: Kept all outliers - they represent large parties with reasonable tip percentages

Key Insight: Outliers aren't always errors! Verify with context (tip percentage in this case).

🔬 PHASE 2: BIVARIATE ANALYSIS (Investigation 2.2)

Overview: Testing 7 Relationships

For each relationship, I followed the scientific method:

Hypothesis - Make a prediction
Visualization - Create appropriate chart
Analysis - Interpret the pattern
Conclusion - Accept or reject hypothesis

2.1 Relationship #1: Total Bill → Tip

My Hypothesis:

"As total_bill increases, tip will increase WEAKLY"
Reasoning: "Tip is 'keep the change' - not percentage based"
Confidence: MEDIUM
Expected: Weak/no relationship

What I Did:

plt.scatter(data_clean['total_bill'], data_clean['tip'])
plt.xlabel('Total Bill ($)')
plt.ylabel('Tip ($)')
plt.title('Relationship Between Total Bill and Tip Amount')
plt.show()

Results:

Pattern: Strong upward linear trend
Correlation: r = 0.67 (Strong positive)
Points tightly clustered around imaginary line

Hypothesis Verdict: ❌ REJECTED

What I Learned:

My hypothesis was WRONG - and that's okay!
Reality: People tip 15-20% of bill (percentage-based, not "keep change")
Mechanism: Bill × 15-20% = Tip (mathematical relationship)
Key insight: "Learning happens with mistakes" - being wrong is part of science!

Business Insight: Higher bills = higher tips. Restaurants should encourage higher spending.

2.2 Relationship #2: Party Size → Total Bill

My Hypothesis:

"As party size increases, total_bill increases STRONGLY"
Reasoning: "More people = more food (obvious!)"
Confidence: HIGH

What I Did:

plt.scatter(data_clean['size'], data_clean['total_bill'])

Results:

Pattern: Grouped upward trend (vertical columns)
Correlation: r = 0.60 (Medium-strong positive)
Party size = discrete (1,2,3,4,5,6), not continuous
Size 2 most common, with widest bill range ($10-$40)

Hypothesis Verdict: ✅ CONFIRMED (but weaker than expected)

Key Insight:

"Party size predicts bill, but doesn't determine it completely"
A couple can outspend a group of 4 depending on what they order
What people ORDER matters more than HOW MANY people

2.3 Relationship #3: Party Size → Tip

My Hypothesis:

"As party size increases, tip increases STRONGLY"
Reasoning: "More people → bigger bill → percentage-based tip → more tip"
Confidence: HIGH

Results:

Pattern: Upward trend from size 1-4, then FLATTENS at 5-6
Correlation: r = 0.49 (Medium-weak)
Non-linear relationship!

Hypothesis Verdict: ⚠️ PARTIALLY CORRECT

Surprising Discovery:

Tips increase up to party size 4
Tips PLATEAU at sizes 5-6 (don't increase further!)

Possible Explanations:

Automatic gratuity - Restaurants add mandatory 15-18% for large parties
Social loafing - "Someone else will tip well, so I don't need to"
Different occasions - Large parties = kids/families (tip standard)
Splitting complications - Harder to calculate when splitting 6 ways

Key Insight: Large parties tip differently than expected - real behavioral economics!

2.4 Relationship #4: Day of Week → Tip

My Hypothesis:

Highest: Sunday (weekend celebration mood)
Lowest: Wednesday (people just filling stomach)
Expected difference: MEDIUM
Confidence: MEDIUM

What I Did:

sns.boxplot(x='day', y='tip', data=data_clean,
            order=['Sun','Mon','Tue','Wed','Thur','Fri','Sat'])

Results:
| Day | Avg Tip | Verdict |
|-----------|---------|----------------|
| Saturday | $3.00 | 🏆 Highest |
| Sunday | $2.90 | High |
| Mon/Tue/Wed| $2.25 | 🔻 Lowest (tie)|

Hypothesis Verdict: ⚠️ PARTIALLY WRONG

What I Got Wrong:

Predicted Sunday highest → Actually Saturday highest
Predicted Wednesday lowest → Correct (tied with Mon/Tue)

Key Observations:

Saturday has most high-tip outliers (special occasions, date nights)
Sunday has LARGEST box (most variation) - diverse crowd
Weekdays cluster together (consistent lower tipping)

Key Insight:
"Sunday = diverse people = diverse tipping = large variation in tips"

2.5 Relationship #5: Time (Lunch vs Dinner) → Tip

My Hypothesis:

Dinner will have higher tips
Reasoning: "Night time = people more generous; lunch = people in rush"
Confidence: MEDIUM

Results:
| Time | Avg Tip | Difference |
|--------|---------|------------|
| Dinner | $3.00 | — |
| Lunch | $2.20 | -$0.80 |

Hypothesis Verdict: ✅ CONFIRMED!

Key Insight:

$0.80 difference - this is the BIGGEST categorical effect!
Time of day is the STRONGEST categorical predictor
Lunch customers are rushed, less satisfied with service
Dinner is relaxed, celebratory atmosphere

Business Recommendation: Prioritize dinner service quality!

2.6 Relationship #6: Sex (Male vs Female) → Tip

My Hypothesis:

Males will tip MORE
Reasoning: "Female waitresses + male customers trying to impress"
Confidence: MEDIUM

Results:
| Sex | Avg Tip | Difference |
|--------|---------|------------|
| Female | $3.20 | — |
| Male | $3.00 | -$0.20 |

Hypothesis Verdict: ❌ REJECTED!

What I Got Wrong:

Females actually tip SLIGHTLY more (or it's basically equal)
The difference is minimal ($0.20)
Sex is NOT a strong predictor

Key Insight: Gender stereotypes about tipping don't hold up in data!

2.7 Relationship #7: Smoker vs Non-Smoker → Tip

My Hypothesis:

Non-smokers will tip MORE
Reasoning: "Smokers save money for cigarettes instead of tipping"
Confidence: LOW

Results:
| Smoker Status | Avg Tip | Difference |
|---------------|---------|------------|
| Smokers | $3.00 | — |
| Non-smokers | $2.80 | -$0.20 |

Hypothesis Verdict: ❌ REJECTED!

Honest Reflection: "Cannot figure out why" - and that's okay!

Possible Explanations:

Smokers sit outside/at bar (different atmosphere?)
Correlation, not causation (maybe age/demographic differences)
Small difference ($0.20) might be random chance
Need more data to understand

Key Insight: Not every pattern has an obvious explanation - intellectual honesty matters!

📈 PHASE 3: CORRELATION ANALYSIS

What I Did:

# Correlation matrix
correlation_matrix = data_clean[['total_bill', 'tip', 'size']].corr()
sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm')

Results:

Pair	Correlation	Strength	Interpretation
total_bill ↔ tip	0.67	Strong	🏆 Strongest predictor
size ↔ total_bill	0.60	Medium-Strong	More people = more food
size ↔ tip	0.49	Medium-Weak	Non-linear (plateaus)

Key Insight:
"Tip percentage is fixed as that of total bill" - this explains the 0.67 correlation perfectly!

🎨 PHASE 4: PAIRPLOT (Visual Summary)

What I Did:

sns.pairplot(data_clean, 
             vars=['total_bill', 'tip', 'size'],
             hue='time',  # Color by lunch/dinner
             diag_kind='hist')

Observations:

From diagonal (distributions):

Most common tip: ~$2-3
Most common bill: ~$15-20
Most common party size: 2 people

From scatter plots:

total_bill vs tip: Clear upward trend (confirms r=0.67)
size vs others: Grouped patterns (discrete variable)
Lunch (blue) vs Dinner (orange): Overlap mostly, but dinner shifts slightly higher

Overall Impression: Relationships are SOMEWHAT CLEAR - not perfect, but strong enough to be meaningful

🎯 KEY FINDINGS SUMMARY

Strongest Predictors (Ranked):

Total Bill (r=0.67) 🥇
- Explains ~45% of tip variation (r² = 0.67² = 0.45)
- Clear linear relationship
- Percentage-based tipping (15-20%)
Time of Day ($0.80 difference) 🥈
- Dinner tips $0.80 more than lunch
- Strongest categorical effect
- Reflects rushed vs relaxed dining
Party Size (r=0.49 with tip) 🥉
- Medium effect, but NON-LINEAR
- Plateaus at size 5-6
- Different behavior for large groups
Day of Week ($0.75 difference)
- Saturday highest ($3.00)
- Weekdays lowest (~$2.25)
- Weekend vs weekday effect
Sex ($0.20 difference) - WEAK
- Minimal difference
- Nearly equal tipping
Smoker Status ($0.20 difference) - WEAK
- Minimal difference
- Unexplained pattern

💡 BUSINESS RECOMMENDATIONS

Based on data analysis, restaurant owners should:

1. FOCUS ON INCREASING BILL AMOUNT 🎯

Why: Strongest correlation (0.67) - higher bills directly lead to higher tips

Actions:

Upsell appetizers, drinks, desserts
Create combo deals that increase bill
Train servers on suggestive selling
Offer premium menu items

Expected Impact: 10% increase in average bill → ~10% increase in tips

2. PRIORITIZE DINNER SERVICE 🌙

Why: Dinner tips $0.80 (36%) more than lunch

Actions:

Allocate best servers to dinner shift
Focus marketing on dinner hours
Create special dinner ambiance
Dinner-specific promotions

Expected Impact: Shift focus to higher-margin time period

3. OPTIMIZE FOR PARTY SIZES 2-4 👥

Why: These sizes have best tip-to-effort ratio

Actions:

Table arrangements favor 2-4 person parties
Special deals for couples/small groups
Don't overinvest in large party accommodations (tips plateau)

Expected Impact: Maximize tips per table/server time

4. WEEKEND FOCUS 📅

Why: Saturday/Sunday have higher tips

Actions:

Premium staffing on weekends
Weekend specials/events
Higher-end menu items on weekends

5. DON'T DISCRIMINATE BY SEX/SMOKER ⚖️

Why: These factors have minimal effect

Insight: Treat all customers equally - demographics don't significantly predict tipping

🧠 PERSONAL LEARNING JOURNEY

What I Learned About Data Science:

1. The Scientific Method Works!

Make hypothesis → Test → Analyze → Conclude
Being wrong is GOOD - that's how we learn!
Quote: "Learning happens with mistakes"

2. Hypotheses Can Be Wrong

My Wrong Predictions:

❌ Thought tipping was "keep the change" → Actually percentage-based
❌ Thought Sunday would have highest tips → Actually Saturday
❌ Thought males tip more → Actually nearly equal
❌ Thought smokers tip less → Actually slightly more

Lesson: Don't trust assumptions - test with data!

3. Correlation ≠ Causation

Smokers tip more, but WHY?
Could be confounding variables (age, location, etc.)
Need more data to understand mechanisms

4. Context Matters

Outliers aren't always errors
$10 tip on $50 bill = 20% (normal!)
Always calculate percentages/ratios for context

5. Data Quality First

Clean data = reliable analysis
Check for: missing values, duplicates, outliers
This dataset was excellent (0 missing!)

6. Visualization is Powerful

Scatter plots → see relationships
Box plots → compare groups
Correlation matrix → see everything at once
Pairplot → ultimate summary

7. Different Charts for Different Data

Numerical vs Numerical → Scatter plot
Categorical vs Numerical → Box plot / Bar chart
All at once → Pairplot, Correlation matrix

What I Learned About Python/Tools:

Python Libraries:

import pandas as pd           # Data manipulation
import numpy as np            # Numerical operations
import matplotlib.pyplot as plt  # Basic plotting
import seaborn as sns         # Statistical plotting

Key Functions Mastered:

Pandas:

data.head()              # First 5 rows
data.shape               # Dimensions
data.describe()          # Statistics
data.isnull().sum()      # Missing values
data.duplicated().sum()  # Duplicates
data.drop_duplicates()   # Remove duplicates
data['column']           # Select column
data[condition]          # Filter rows
data.groupby().mean()    # Group and aggregate
data.corr()              # Correlation matrix

Matplotlib:

plt.scatter(x, y)        # Scatter plot
plt.xlabel()             # X-axis label
plt.ylabel()             # Y-axis label
plt.title()              # Title
plt.grid()               # Grid lines
plt.show()               # Display plot

Seaborn:

sns.boxplot(x='category', y='number', data=df)
sns.heatmap(corr_matrix, annot=True)
sns.pairplot(df, vars=['col1','col2'], hue='category')

Important Concepts:

1. Order Matters!

# WRONG
outliers = data[data['tip'] > 5]
data['tip_pct'] = data['tip'] / data['total_bill']
print(outliers['tip_pct'])  # ERROR!

# RIGHT
data['tip_pct'] = data['tip'] / data['total_bill']
outliers = data[data['tip'] > 5]
print(outliers['tip_pct'])  # WORKS!

2. Matplotlib vs Seaborn:

matplotlib = Low-level, flexible, more code
seaborn = High-level, easy, pretty defaults
Use both together!

3. Data Types Matter:

float64 / int64 = Can do math
object = Text, can't do math

Skills I Developed:

✅ Technical Skills:

Data cleaning and preprocessing
Exploratory data analysis
Statistical thinking
Data visualization
Python programming
Using Jupyter notebooks

✅ Analytical Skills:

Hypothesis formation
Pattern recognition
Critical thinking
Drawing insights from data
Making business recommendations

✅ Soft Skills:

Scientific method application
Intellectual honesty ("I don't know")
Learning from mistakes
Persistence through challenges
Clear communication of findings

📊 COMPLETE VISUALIZATIONS CREATED

✅ Scatter Plot: total_bill vs tip
✅ Scatter Plot: size vs total_bill
✅ Scatter Plot: size vs tip
✅ Box Plot: tip by day of week
✅ Box Plot: tip by time (lunch/dinner)
✅ Box Plot: tip by sex
✅ Box Plot: tip by smoker status
✅ Correlation Matrix Heatmap
✅ Pairplot (all relationships)

Total: 9 professional visualizations

🎓 FINAL REFLECTION

What Worked Well:

Systematic approach (hypothesis → test → analyze)
Using appropriate visualizations for each relationship
Being open to being wrong
Thorough data cleaning before analysis

What I'd Do Differently:

Could explore interaction effects (e.g., day × time)
Could calculate tip percentages earlier for context
Could test non-linear relationships more formally

Most Surprising Finding:

"Party size plateaus at 5-6 people!"

Expected linear relationship
Discovered real-world behavioral economics
Shows the value of looking at data, not just assumptions

Most Important Lesson:

"Total amount of spending is the determining factor"

Simple but powerful
Actionable for businesses
Data-driven decision making

End of Journey Summary

"The only real mistake is the one from which we learn nothing." - Henry Ford

"In God we trust. All others must bring data." - W. Edwards Deming

नियम नियति (Niyam Niyati) - Youth Bill Engagement Platform

NIRANJAN LAMICHHANE — Fri, 05 Dec 2025 15:03:38 +0000

📋 Project Overview

Niyam Niyati (meaning "Laws & Destiny" in Nepali) is an AI-powered civic engagement platform that helps Nepali youth (ages 16-25) understand complex legislative bills through empathy-driven persona exploration. The application transforms abstract legal documents into relatable human stories, enabling young citizens to understand legislation through the eyes of those it affects most.

The Problem We're Solving

Young people often feel disconnected from legislative processes because:

Legal language is intimidating and inaccessible
Bills seem abstract with no clear personal relevance
There's no easy way to understand how laws affect different people
Traditional civic education fails to engage digital-native youth

Our Solution

We built a Frankenstein monster of technologies - stitching together AI persona generation, empathy-building mechanics, and civic engagement into something unexpectedly powerful. The platform:

Presents real Nepali legislative bills in a digestible format
Generates AI-powered personas representing diverse demographics affected by each bill
Shows concrete scenarios demonstrating bill impacts (action → consequence)
Enables perspective-based voting from each persona's viewpoint
Provides AI-generated insights and recommendations for both youth and lawmakers

🔧 Technologies Stitched Together (The Frankenstein Factor)

Our app is a true chimera, combining seemingly incompatible elements:

Technology	Purpose	The "Unexpected" Element
Perplexity AI (Sonar)	Persona & insight generation	Using search-grounded AI for legislative context
React 18 + Vite	Frontend framework	Spec-driven development with Kiro
Tailwind CSS	Styling	Fully responsive civic engagement UI
LocalStorage	Data persistence	Client-side vote aggregation
React Router v6	Navigation	Multi-perspective user journey

The Chimera Effect

We stitched together:

AI + Civic Education: Using LLMs to make legislation accessible
Empathy Mechanics + Voting: Understanding others' perspectives before casting votes
Youth Culture + Formal Legislation: Making legal documents relatable to Gen-Z
Nepali Context + Global Tech: Localizing AI for a specific cultural context

🛠️ How Kiro Was Used

Spec-Driven Development

We leveraged Kiro's spec-driven approach extensively. Our .kiro/specs/youth-bill-engagement/ directory contains:

`requirements.md` - Detailed User Stories & Acceptance Criteria

### Requirement 1
**User Story:** As a user, I want to view and select from available bills, 
so that I can explore legislation that interests me.

#### Acceptance Criteria
1. WHEN the System starts THEN the System SHALL display four bill cards...
2. WHEN a user hovers over a bill card THEN the System SHALL apply animation...

This structured approach ensured every feature was traceable to a user need.

`design.md` - Architecture & Correctness Properties

We defined 21 correctness properties that served as formal specifications:

### Property 4: API response validation
*For any* API response, the validation function should verify that the JSON 
structure contains persona_name, persona_profile, what_stays_same (4 items), 
what_becomes_riskier (5 items), and scenarios (4 items with action and 
consequence fields).

`tasks.md` - Implementation Plan

Kiro helped break down the project into actionable tasks with clear dependencies and requirement mappings.

Steering Documents

Our .kiro/steering/ directory contains:

product.md - Product vision and core features
tech.md - Technology stack and commands
structure.md - Project architecture and naming conventions

How Kiro Improved Development

Structured Conversations: We used Kiro to discuss requirements before coding, resulting in clearer implementations
Spec-to-Code Traceability: Every component includes requirement references:

   /**
    * Persona Screen Component
    * Requirements: 3.1, 3.2, 3.3, 3.4, 3.5, 4.1, 4.2, 4.3, 4.4, 6.4
    */

Property-Based Testing Strategy: Kiro helped define properties for testing:

   // Feature: youth-bill-engagement, Property 5: Persona data storage round-trip

Consistent Architecture: Steering docs ensured consistent patterns across the codebase

Most Impressive Code Generation

Kiro helped generate our Perplexity API service with sophisticated prompt engineering:

const prompt = `You are helping Nepali youth (ages 16-25) understand legislation. 
Generate 4 diverse personas affected by this bill.

Requirements:
- Use neutral, unbiased tone
- Use youth-friendly language appropriate for ages 16-25
- Provide relatable context for Nepali youth
- Each persona should represent a different demographic

Return ONLY valid JSON in this exact structure:
[{
  "persona_name": "Name (Age, Occupation/Role)",
  "persona_profile": "2-3 sentence description...",
  "what_stays_same": ["bullet 1", "bullet 2", ...],
  "what_becomes_riskier": ["bullet 1", "bullet 2", ...],
  "scenarios": [{ "action": "...", "consequence": "..." }]
}]`;

The spec-driven approach ensured the AI output matched our exact requirements.

🎯 Key Features

1. Bill Selection Screen

Displays 4 real Nepali legislative bills with emojis and summaries
Hover animations for engaging UX
Loading states during AI persona generation

2. Persona Exploration

Each bill generates 4 AI personas with:

Persona Profile: Who they are and their situation
What Stays the Same: 4 aspects unchanged by the bill
What Becomes Riskier: 5 potential risks from the bill
Scenarios: 4 action-consequence pairs

3. Perspective-Based Voting

Vote from each persona's perspective (Support/Oppose/Unsure)
Optional message to lawmakers
Votes stored locally for aggregation

4. Community Insights

Aggregated voting statistics per persona
AI-generated summary and key insights
Recommendations for both youth and lawmakers

📱 Bills Included

Bill	Description
📱 Social Media Act (Bill), 2081	Regulates social media platforms and users
🔒 IT and Cyber Security Bill, 2082	Defines cybercrime offenses and penalties
🎓 School Education Bill, 2080	Restructures school levels and standards
🔐 Privacy Act, 2075	Governs personal data protection

🚀 Running the Application

Prerequisites

Node.js v16+
Perplexity API key

Installation

npm install

Configuration

Create .env.local:

VITE_PERPLEXITY_API_KEY=your_api_key_here

Development

npm run dev

Testing

npm run test

🧪 Testing Coverage

✅ 34 automated tests passing
✅ Storage service tests (17 tests)
✅ Persona screen tests (7 tests)
✅ Application structure tests (10 tests)

🎨 Design Philosophy

Neutral Presentation

The platform maintains strict neutrality - presenting facts without editorial bias, allowing users to form their own opinions.

Empathy-First Approach

By experiencing legislation through diverse personas, users develop empathy for how laws affect different people.

Youth-Friendly Language

All AI-generated content is tailored for ages 16-25 with relatable context and accessible language.

Responsive Design

Desktop (≥1024px): Two-column layouts
Mobile (<1024px): Single-column optimized views

🏆 Why This Fits "Frankenstein"

Like Dr. Frankenstein's creation, we've stitched together parts that shouldn't work together:

The Brain (AI): Perplexity's Sonar model provides intelligent persona generation
The Heart (Empathy): Perspective-taking mechanics create emotional connection
The Body (Tech Stack): Modern web technologies give it form
The Soul (Purpose): Civic engagement gives it meaning
The Spark (Kiro): Spec-driven development brought it to life

The result is something greater than the sum of its parts - a platform that makes democracy accessible to the next generation.

📜 License

Open Source under MIT License

🔗 Links

Repository: [https://github.com/NiranjanNlc/niyam-niyati]
Live Demo: [https://splendid-paletas-a20883.netlify.app/]
Demo Video: [https://www.youtube.com/watch?v=L3PMRx3ALEo]

👥 Team

Built with ❤️ for Nepali youth and democracy

"It's alive!" - Dr. Frankenstein (and us, when the specs finally compiled)

Kiro helped me to help people to escape tutorial hell

NIRANJAN LAMICHHANE — Mon, 15 Sep 2025 13:52:06 +0000

Many developer , who are in the stage of the learning and beginner level , get stuck in “tutorial hell”—watching endless tutorials . This contributes to the feeling of passive learning and growth while in real , one has not learned and grown at all .

Kiro as a planner buddy

We used Kiro to create this app to engage the developer and learner in the active project based learning by taking action and building project so that real growth and learning can be achieved .

We use kiro especially for the specification driven desighn ,

After giving the prompt for the project , it generated the requirement.md on which there are many unnecessary requirement , which we do not want in the first place . We cut down them and move into desighn phase .

Kiro creates desighn.md which is especially for full fledged full stack project by using react , typescript and nodejs . We prompt kiro to use the hardcoded instruction and just html,css and js for the project and it updates the desighn documentaion accordingly .

Kiro produces tasks based on the desighn.md and requirement.md . We prompt kiro to make tasks just concise to make beautiful ui and serves the purpose of MVP.

Kiro produces the requirement.md and desighn.md file while building from scratch .

By prompting kiro to change and update those file according to our requirement helps us to make complete control of the project without losing the vibe of vibe buildong or vibe coding.

Delegating Version control tasks to kiro Hook agent

We automate the version control in the project with the help of kiro project .

We prompt kiro to create the hook that gets activated when the tasks is completed in tasks.md file and commit in the project with the message of tasks .

Kiro Main helping hand

Kiro produces the requirement.md and desighn.md file while building from scratch .

By prompting kiro to change and update those file according to our requirement helps us to make complete control of the project without losing the vibe of vibe buildong or vibe coding

Why I never returned back to windows after using linux ?

NIRANJAN LAMICHHANE — Fri, 20 May 2022 06:35:57 +0000

During the first year of my engineering , my laptop hard drive crashed and I need to get a new one .
Along with the hard drive, i need to think about OS I gonna use on it . Linux was a short of hype among my peer circle. I thought of why not give a try .
After that , I was so much in deep love with linux that I wonder why the people are still stick around windows.

Here are some of the reason why I never returned back to windows after using linux :

Its not that much hard . All of us assume that linux is all about command line but its not so . I tried Ubuntu and its GUI is more appealing and easy to use than that of Windows. Remember , linux provides both command line and graphical user interface.

Easy to install software: In windows , in order to install software , you need to download .exe file from internet , run it as a administrator and perform other series of steps. However, I can install a software on ubuntu in a single command .

Faster and smooth :The smoothness and fastness of the ubuntu is felt by me while running Android studio. My PC almost freezes while running android studio in Windows , however I have not encountered such isuses till now in linux.

Free software : The open source alternatives of microsoft office is Libre Office in Ubuntu, which don't incur any cost for usage.

Community :Linux has large community and fanbase. If we faces the problem in linux , the solution is just a one thread away in our browser.

Pride : Using linux gave me a sense of entitlement and made me a proud.

What makes you hooked in linux and make you away from windows?

DEV Community: NIRANJAN LAMICHHANE

Coffee, Code, and Junie: My Journey to 10x Productivity with JetBrains AI

Dream of 10x Developer :

My Favourite part :

Important Contribution :

Impressive Refactoring :

UI Magic for Backend Developers:

The Flow State Achievement:

Most Misused Capability :

Wish it was there :

My Data Science Journey: Restaurant Tips Analysis

📊 PROJECT OVERVIEW

Dataset Information

Project Goal

🧹 PHASE 1: DATA CLEANING (Investigation 1.3)

1.1 Missing Values Investigation

1.2 Duplicate Detection

1.3 Outlier Investigation

🔬 PHASE 2: BIVARIATE ANALYSIS (Investigation 2.2)

Overview: Testing 7 Relationships

2.1 Relationship #1: Total Bill → Tip

2.2 Relationship #2: Party Size → Total Bill

2.3 Relationship #3: Party Size → Tip

2.4 Relationship #4: Day of Week → Tip

2.5 Relationship #5: Time (Lunch vs Dinner) → Tip

2.6 Relationship #6: Sex (Male vs Female) → Tip

2.7 Relationship #7: Smoker vs Non-Smoker → Tip

📈 PHASE 3: CORRELATION ANALYSIS

What I Did:

Results:

🎨 PHASE 4: PAIRPLOT (Visual Summary)

What I Did:

Observations:

🎯 KEY FINDINGS SUMMARY

Strongest Predictors (Ranked):

💡 BUSINESS RECOMMENDATIONS

1. FOCUS ON INCREASING BILL AMOUNT 🎯

2. PRIORITIZE DINNER SERVICE 🌙

3. OPTIMIZE FOR PARTY SIZES 2-4 👥

4. WEEKEND FOCUS 📅

5. DON'T DISCRIMINATE BY SEX/SMOKER ⚖️

🧠 PERSONAL LEARNING JOURNEY

What I Learned About Data Science:

1. The Scientific Method Works!

2. Hypotheses Can Be Wrong

3. Correlation ≠ Causation

4. Context Matters

5. Data Quality First

6. Visualization is Powerful

7. Different Charts for Different Data

What I Learned About Python/Tools:

Python Libraries:

Key Functions Mastered:

Important Concepts:

Skills I Developed:

📊 COMPLETE VISUALIZATIONS CREATED

🎓 FINAL REFLECTION

What Worked Well:

What I'd Do Differently:

Most Surprising Finding:

Most Important Lesson:

नियम नियति (Niyam Niyati) - Youth Bill Engagement Platform

📋 Project Overview

The Problem We're Solving

Our Solution

🔧 Technologies Stitched Together (The Frankenstein Factor)

The Chimera Effect

🛠️ How Kiro Was Used

Spec-Driven Development

requirements.md - Detailed User Stories & Acceptance Criteria

design.md - Architecture & Correctness Properties

tasks.md - Implementation Plan

Steering Documents

How Kiro Improved Development

Most Impressive Code Generation

🎯 Key Features

1. Bill Selection Screen

2. Persona Exploration

3. Perspective-Based Voting

4. Community Insights

`requirements.md` - Detailed User Stories & Acceptance Criteria

`design.md` - Architecture & Correctness Properties

`tasks.md` - Implementation Plan