DEV Community: Akshay Verma

Designing offline-first when there's no server

Akshay Verma — Sun, 20 Sep 2026 21:16:14 +0000

I'm building an app with no backend. No accounts, no sync, no API. The database is a SQLite file on the phone, and that's the whole system.

That sounds simpler than a client/server app. In some ways it is. But removing the server doesn't remove the hard problems — it relocates them.

The device is the source of truth

With no server, there's no authority to reconcile against and no "refresh from the backend" escape hatch. If the local data is wrong, it's wrong permanently. Every bug becomes a data-integrity bug.

This changes how you treat writes. You stop thinking "I'll fix it in a migration later" and start thinking about what the row should have been in the first place.

The mistake I made: recomputing history

The app summarises a period, and the summary is derived from underlying records. Naturally, I computed it on demand.

Then I edited a figure that had been recorded months earlier — a correction to a current value. Every past period silently changed. Records I'd already finalised now showed different numbers, because they were being recalculated from data that had since moved.

Nothing crashed. There was no error. History just quietly rewrote itself, and I only noticed because a total I remembered was different.

The fix: freeze at the boundary

When a period closes, compute the summary once and store it. From then on, that period is read, never recalculated:

const plan = month.closed
? readSnapshot(month.period) // frozen at close
: computePlan(month); // still live, safe to recompute

One rule, enforced everywhere: a closed period is read from its snapshot, never recomputed. Any screen showing those figures checks closed before letting anything change.

This is just event-sourcing intuition applied small. Derived data is fine while its inputs are still in play. The moment a period is done, the derivation becomes a fact and needs storing like one.

Backup, and the parts that shouldn't come back

No server also means no backup unless you write one. Mine dumps every table to a file the user controls.

Two things that weren't obvious:

Keep the deleted rows. Soft-deleted records stay in the dump. Restoring a backup that silently resurrects things the user deleted is worse than not restoring at all.

Not every setting is data. Theme and language belong to the device, not the backup. Restoring on a new phone shouldn't drag the old phone's dark-mode preference along. Everything else restores; those two are deliberately skipped.

The failure mode to design against

Here's the one that will bite you: add a new table, forget to add it to the dump-and-restore routine, and it vanishes on restore. No error. The app works fine — the data is just gone.

I don't have a clever solution, only a blunt one: the backup routine has a single explicit list of tables, and adding a table means touching that list in the same commit. Making the omission visible in code review is the whole defence.

Worth it?

Yes — for this app. No server means no hosting, no auth, no breach surface, no subscription to keep the lights on, and it works on a train.

But "offline-first" isn't "backend-free and therefore easy". You're still running a database, still versioning a schema, still writing backup and restore. You've just moved all of it onto a device you don't control, where the user can uninstall your entire production environment by accident.

Why my builds don't run on my laptop

Akshay Verma — Sun, 20 Sep 2026 21:15:09 +0000

My React Native app has never been compiled on my own machine. Not once. That started as a limitation and turned into the thing that keeps my releases boring.

The setup

During development I don't build at all. The JavaScript layer runs in Expo Go on a physical phone over the local network — save a file, see it on the device a second later. That covers the overwhelming majority of what I write: screens, state, business logic, database queries.

Native code is different. The moment you add a library with native modules, or touch the Android manifest, you need a real build. That build goes to EAS, not to my laptop.

Why not locally?

Because "works on my machine" is a sentence about my machine, not my app.

A local Android build depends on which JDK I happen to have, which SDK platforms are installed, which NDK version, which Gradle cache. None of that lives in my repo. If it breaks in six months, I'm debugging my laptop's history, not my code.

A hosted build is defined by files that are committed: app.json, eas.json, package.json. The build environment is an input, not an accident. That's the whole argument, and it's the same argument as any other CI.

The quota, which turned out to be a feature

My build credits reset monthly, and there aren't many. I can't rebuild on every commit — realistically I build every two or three days.

I expected this to be pure friction. Instead it changed how I work. When a build is free you use it as a test. When it's scarce you push verification earlier:

npx tsc --noEmit # types, across the whole project
npx expo export --platform android # does the bundle actually build?

Both run in seconds, locally, and catch most of what would otherwise fail ten minutes into a remote build. Nothing goes to EAS until both are clean. I've spent far fewer builds on "oh, a typo" than I used to.

The constraint also forces batching. Instead of one native change per build, I queue several, then verify them together on the device. Fewer builds, longer device-testing sessions, and honestly better testing — because I'm looking at a batch of changes with fresh eyes rather than confirming the one thing I just wrote.

The config trap

One hard-won detail. I needed Android package visibility so the app could detect whether certain other apps were installed. The obvious move is to add a queries block to app.json.

It does nothing. It isn't an error, there's no warning — the key is simply ignored, and you find out when the feature silently fails on a real device.

The actual mechanism is a config plugin that modifies the manifest during prebuild:

const { withAndroidManifest } = require('expo/config-plugins');

module.exports = (config) =>
withAndroidManifest(config, (cfg) => {
const manifest = cfg.modResults.manifest;
manifest.queries = [{ package: [{ $: { 'android:name': 'com.example.target' } }] }];
return cfg;
});

The lesson generalises past Expo: when a config key is silently ignored, you don't have a configuration problem, you have a wrong layer problem. Find the layer that actually owns the output.

What I'd tell past me

Don't fight to get native builds working locally. The reproducibility you get from a defined build environment is worth more than the iteration speed you think you're losing — because the fast loop was never the native build anyway. It was Expo Go, and that still runs on your desk.

Stop storing money as a float

Akshay Verma — Sun, 20 Sep 2026 20:51:04 +0000

I'm building an expense-splitting app for couples. Early on, I stored every amount as a double. It worked until two people split a ₹1,000.05 grocery bill three ways — and the app insisted someone still owed ₹0.01 that nobody could pay.

Why it happens

0.1 + 0.2 === 0.30000000000000004. Binary floating point can't represent most decimal fractions exactly, the same way decimal can't write 1/3. Individually the error is invisible. Across a month of bills, summed and re-split, it surfaces as a phantom rupee.

The fix: integers, always

Store the smallest unit — paise, not rupees:

const toPaise = (rupees) => Math.round(rupees * 100);
const format = (paise) => ₹${(paise / 100).toFixed(2)};

Every amount in the database is an integer. Addition and subtraction are now exact. You convert to a display string at the very last moment, and never compute on that string.

Splitting without losing a paisa

Division is where money actually disappears. ₹1000.05 three ways is 33335 paise each with 0 left, but ₹1000.04 leaves a remainder of 1. Rounding each share independently either invents or destroys money.

Distribute the remainder explicitly instead:

function split(totalPaise, weights) {
const sum = weights.reduce((a, b) => a + b, 0);
const shares = weights.map(w => Math.floor(totalPaise * w / sum));
let left = totalPaise - shares.reduce((a, b) => a + b, 0);
for (let i = 0; left > 0; i = (i + 1) % shares.length, left--) shares[i]++;
return shares;
}

The shares now always sum to exactly the total. Someone absorbs an extra paisa — but it's deliberate, and it's visible.

The rule

Integers for storage and arithmetic. Strings only for display. Never a float, and never a rounded intermediate value.

The bug that started this took an evening to find and three lines to fix. Money is the one place where "close enough" is a bug report.

AI: A Double-Edged Sword, Shaping Our Future

Akshay Verma — Sat, 07 Dec 2024 20:03:14 +0000

Artificial intelligence (AI) is no longer a futuristic concept but a tangible reality that's reshaping industries and our everyday lives. Let's delve into the remarkable capabilities of AI today, drawing inspiration from recent news and advancements.

AI's Evolving Prowess

AI has progressed from a niche field to a mainstream technology, thanks to breakthroughs in machine learning and deep learning. Here's a glimpse into some of its remarkable achievements:

Natural Language Processing (NLP):
AI can now understand, interpret, and generate human language with remarkable accuracy. This has led to advancements in chatbots, virtual assistants, and language translation tools. For instance, OpenAI's ChatGPT has been making headlines for its ability to engage in informative and comprehensive conversations.
Computer Vision:
AI algorithms can analyze and interpret visual information from images and videos. This has applications in autonomous vehicles, medical image analysis, and surveillance systems. Recent news highlights the success of AI-powered systems in detecting and diagnosing diseases like cancer with greater accuracy than human experts.
Generative AI:
AI models can create new content, such as text, images, and even music. This has opened up exciting possibilities in creative fields like art, design, and content creation. Tools like Midjourney and Stable Diffusion are generating stunningly realistic images from simple text prompts.
AI in Healthcare:
AI is revolutionizing healthcare by assisting in drug discovery, personalized medicine, and medical image analysis. AI-powered systems are helping researchers identify potential drug candidates and develop more effective treatments.
AI in Finance:
AI is being used to detect fraud, analyze market trends, and automate financial processes. AI-powered trading algorithms are making split-second decisions to maximize returns.
AI in Education:
AI-powered tutoring systems are providing personalized learning experiences to students. AI can also automate administrative tasks, freeing up educators to focus on teaching.

Beyond the Headlines

While these advancements are awe-inspiring, it's crucial to remember that AI is still a developing field. There are ongoing challenges to address, such as ensuring AI's ethical development and mitigating potential biases. However, the future of AI holds immense promise, with the potential to solve some of humanity's most pressing problems.

The Road Ahead

As AI continues to evolve, we can expect even more groundbreaking applications. From AI-powered robots assisting in disaster relief to AI-driven climate change solutions, the possibilities are endless. However, it is essential to approach AI development with caution and ensure that it benefits society as a whole.

In conclusion, AI is no longer a distant dream but a powerful tool that is transforming our world. By understanding its capabilities and limitations, we can harness its potential for good and shape a future where AI serves as a force for positive change.

Demystifying AI: A Beginner's Guide to Artificial Intelligence

Akshay Verma — Sat, 29 Jun 2024 09:49:29 +0000

Artificial Intelligence (AI) often feels like a buzzword that’s always floating around, yet its true essence remains elusive to many. Whether it’s the futuristic robots in movies or the smart assistants on our phones, AI seems to be everywhere. But what is AI really? In this beginner's guide, we’ll unravel the mysteries of artificial intelligence (AI), explaining it in a way that anyone can understand, without the jargon and complexity.

Table of Contents
SNo. Headings 1 Introduction to AI 2 What is AI? 3 Types of AI 4 Machine Learning and Deep Learning 5 Applications of AI 6 Benefits of AI 7 Challenges of AI 8 The Future of AI 9 AI and Everyday Life 10 Ethics of AI 11 How to Learn About AI 12 Careers in AI 13 Conclusion 14 FAQs

Introduction to AI:
Artificial Intelligence has transitioned from the realm of science fiction to a significant part of our daily lives. From machine learning algorithms recommending what to watch next on Netflix to deep learning powering voice assistants like Siri and Alexa, AI is more integrated into our lives than we might realize. This guide aims to provide a beginner's guide to AI, breaking down the concepts into digestible pieces.

What is AI?
So, what is AI? Simply put, AI is a branch of computer science that aims to create machines that can mimic human intelligence. This includes learning from experiences (like humans do), understanding language, recognizing patterns, and making decisions.

Think of AI as a child learning to recognize objects. Initially, they might not know what a cat is. But over time, by seeing multiple images of cats and receiving feedback, they learn to identify a cat. AI works similarly, using data and algorithms to learn and improve over time.

Types of AI
AI can be broadly categorized into two types:

Narrow AI: This is AI designed to perform a specific task, like facial recognition or internet searches. It’s incredibly effective within its scope but cannot perform tasks outside its designated function.
General AI: This is the type of AI that can perform any intellectual task that a human can. It’s the stuff of science fiction and, as of now, does not exist.
Narrow AI is what we interact with today. For example, the spam filter on your email is a type of narrow AI.

Machine Learning and Deep Learning
To dive deeper into AI, we need to understand two important subsets: machine learning and deep learning.

Machine Learning
Machine learning is a method of data analysis that automates analytical model building. It’s based on the idea that systems can learn from data, identify patterns, and make decisions with minimal human intervention.

For instance, machine learning algorithms can analyze millions of transactions to detect fraudulent activities.

Deep Learning
Deep learning is a subset of machine learning involving neural networks with many layers (hence “deep”). These neural networks try to simulate the way our brains work, which allows the system to learn from large amounts of data.

A practical example of deep learning is image recognition, such as tagging friends in photos on social media platforms.

Applications of AI
AI has a vast range of applications that touch nearly every aspect of our lives. Here are a few key areas:

Healthcare: AI is used for diagnosing diseases, developing treatment plans, and personalizing medicine.
Finance: AI helps in fraud detection, risk management, and automated trading.
Transportation: Self-driving cars and traffic management systems use AI to improve safety and efficiency.
Customer Service: Chatbots and virtual assistants provide 24/7 support, enhancing customer experience.
Entertainment: AI recommends movies, songs, and shows based on your preferences.

Benefits of AI
The benefits of AI are numerous:

Efficiency: AI can perform tasks faster and more accurately than humans.
Cost Savings: Automating routine tasks can save businesses significant amounts of money.
Personalization: AI allows for highly personalized user experiences, such as customized recommendations.
Innovation: AI opens new possibilities in fields like medicine, engineering, and beyond.

Challenges of AI
Despite its many benefits, AI also presents several challenges:

Ethical Concerns: Issues like data privacy, bias in AI systems, and the impact on employment need to be addressed.
Security Risks: AI systems can be vulnerable to hacking and misuse.
Complexity: Developing and maintaining AI systems can be complicated and expensive.
Dependence on Data: AI systems require vast amounts of data, raising concerns about data security and privacy.

The Future of AI
The future of AI holds exciting possibilities. As technology advances, we can expect AI to become even more integrated into our lives. Some potential developments include:

Enhanced Human-AI Collaboration: AI could work alongside humans, augmenting our capabilities and improving productivity.
Advanced Healthcare Solutions: AI could lead to breakthroughs in medical research, diagnosis, and treatment.
Smarter Cities: AI could help manage urban infrastructure, improving the quality of life in cities.
Innovative Products: New AI-powered products and services will continue to emerge, transforming industries.
AI and Everyday Life

AI’s impact on everyday life is profound. Here are some ways AI influences our daily activities:

Voice Assistants: Devices like Google Home and Amazon Echo use AI to assist with tasks.
Smart Homes: AI controls lighting, heating, and security systems for enhanced comfort and safety.
Shopping: AI algorithms power personalized shopping experiences and recommendations.
Navigation: AI improves the accuracy of maps and navigation apps, helping us get to our destinations efficiently.
Health Monitoring: Wearable devices use AI to monitor our health and fitness levels.
Ethics of AI
The ethics of AI is a critical area of discussion. As AI systems become more powerful, ethical considerations become more important. Key ethical issues include:

Bias: Ensuring AI systems are fair and unbiased.
Transparency: Making AI decision-making processes understandable to humans.
Accountability: Determining who is responsible when AI systems fail or cause harm.
Privacy: Protecting personal data and privacy in AI applications.
How to Learn About AI
Learning about AI doesn’t require a technical background. Here are some steps to get started:

Online Courses: Platforms like Coursera, edX, and Udacity offer AI courses for beginners.
Books: There are numerous books on AI that explain concepts in an accessible manner.
Websites and Blogs: Websites like Medium, Towards Data Science, and AI news sites provide valuable insights.
Podcasts and Videos: Many experts share their knowledge through podcasts and YouTube channels.

Careers in AI
AI offers exciting career opportunities across various fields:

Data Scientists: Analyze data and develop AI models.
AI Engineers: Design and build AI systems.
Research Scientists: Conduct research to advance AI technologies.
Product Managers: Oversee the development and deployment of AI products.

Ethicists: Focus on the ethical implications of AI.

Conclusion
Artificial intelligence is transforming the world in ways we are only beginning to understand. From practical applications to ethical considerations, AI presents both incredible opportunities and significant challenges. By demystifying AI, we can better appreciate its impact and potential, preparing ourselves for a future where AI plays an even greater role in our lives.

FAQs

What is AI?
AI, or artificial intelligence, is the simulation of human intelligence in machines that are programmed to think and learn like humans.
How does machine learning differ from deep learning?
Machine learning involves algorithms that allow computers to learn from data, while deep learning uses neural networks with many layers to analyze large amounts of data more effectively.
What are some common applications of AI?
Common applications include healthcare diagnosis, financial fraud detection, self-driving cars, customer service chatbots, and personalized entertainment recommendations.
Are there any ethical concerns associated with AI?
Ethical concerns include data privacy, bias in AI systems, accountability for AI decisions, and the potential impact on jobs and society.
How can someone start learning about AI?
Starting with online courses, reading books, following AI-focused websites and blogs, and listening to podcasts or watching educational videos are good ways to learn about AI.