DEV Community: SandraMeshack

What I Learned in My First Month of Owning a Digital Technology Company

SandraMeshack — Sat, 27 Sep 2025 01:42:35 +0000

Starting a digital technology company has been one of the most exciting—and eye-opening—experiences of my life. It’s been just over a month since I officially launched, and already, I’ve learned more than I could have imagined. From putting myself out there to embracing discomfort, the first 30 days have taught me that entrepreneurship is just as much about personal growth as it is about business strategy.

Here are a few key lessons I’ve taken away so far:

1- You Have to Put Yourself Out There

One of the first and biggest challenges I faced was learning to step outside of my comfort zone. As someone who’s more comfortable behind the screen, the idea of self-promotion didn’t come naturally. But if I’ve learned anything this month, it’s that nobody will know what you do—or how well you do it—unless you’re willing to show up and talk about it.

Whether it's reaching out to potential clients, posting on dev.to, or pitching to strangers, I’ve had to actively put myself in front of people. And the best part? It works. People respond to authenticity and passion.

2- Imposter Syndrome is Real—But Becomes Fuel

There were moments in the first few weeks when I questioned everything. Am I really cut out for this? Do I know enough? Will anyone trust me with their business needs?

Imposter syndrome is real, but I’ve started to use it as fuel instead of a roadblock. Every time I feel unsure, I remind myself: I’ve taken the leap, I’m learning fast, and I’m solving real problems. That alone puts me miles ahead of where I was just a short time ago.

3- Clarity Comes Through Action

I didn’t (and still don’t) have all the answers—but I’ve realised that clarity doesn’t come from planning endlessly. It comes from doing. Every conversation, client meeting, proposal, and late-night brainstorm has helped shape the direction of the business.

Instead of waiting until everything felt “perfect,” I’ve embraced the idea of launching, learning, and refining along the way.

4- Your Network is Everything

In this first month, I’ve leaned heavily on my network—and it’s been incredibly rewarding. From offering advice and introductions to becoming clients and collaborators, people in my circle have shown up in ways I didn’t expect.

I’ve also learned the importance of expanding that network—joining communities, attending industry events, and simply being open to new connections. Building relationships takes time, but every one of them is a potential door to a new opportunity.

5- Systems and Structure Matter (More Than I Thought)

In a digital tech company, it’s easy to focus all your energy on client work or product development. But I quickly realised that without internal systems—things like task management, financial tracking, and communication processes—it’s easy to lose control.

Even as a small team (or solo founder), setting up systems early has helped me stay organised, deliver consistently, and plan ahead with more confidence.

6- Celebrate Small Wins

It’s tempting to always look ahead at what’s next—the next lead, the next project, the next big goal. But I’ve made it a point to pause and celebrate the small wins: launching my website, signing my first client, getting great feedback, or simply making it through a tough week.

These moments are the foundation of what I’m building, and they deserve recognition.

Final Thoughts

The first month of running a digital technology company has been intense, humbling, and deeply rewarding. I’ve learned that entrepreneurship isn’t about having all the answers—it’s about being willing to figure things out as you go, staying curious, and trusting yourself even when it’s hard.

If you’re thinking about starting your own venture, my advice is simple: start before you’re ready, keep learning, and don’t be afraid to put yourself out there. The growth that comes from showing up every day is worth every challenge.

Here’s to month two—and everything that comes next.

You can view my portfolio to know more about me and my services.

Why Every Engineering Team Needs a Technical Writer (Even Small Ones)

SandraMeshack — Sat, 20 Sep 2025 04:16:05 +0000

Ever joined a codebase and spent hours just trying to figure out what talks to what?

Or asked a teammate how something works, only to get a 20-minute Slack voice note that left you more confused than before?

Yeah — documentation matters. And not just the README.md kind.

That’s where a technical writer comes in.

More Than Just "Making Things Pretty"

Most engineering teams don’t realise they need a technical writer until it’s too late — onboarding is painful, tribal knowledge is leaking, and no one really knows how the system works anymore.

The truth?

A good technical writer doesn’t just “write docs.” They reduce friction, amplify developer velocity, and preserve your team’s collective brain.

In this post, I want to share why every engineering team — even small ones — can benefit from having a technical writer. Not to make things look polished, but to make your systems easier to understand, maintain, and grow.

1- They Reduce the Mental Load on Developers

Developers already carry a heavy cognitive load — thinking about tradeoffs, dependencies, naming, scalability, and edge cases.

Without clear documentation, they also have to hold the entire system in their heads, or interrupt teammates to ask how something works.

A technical writer gives your team an external brain — creating reliable, up-to-date documentation that developers can trust instead of relying on memory or Slack threads.

2- They Make Onboarding Way Less Painful

Every new hire asks the same questions:

What does this service do?
Where’s the API spec?
Why did we build it this way?

A technical writer turns those repeated questions into structured, reusable answers.

That means new developers ramp up faster, get blocked less often, and feel more confident sooner — without draining the time of your most senior engineers.

3- They Help Your Team Communicate With Itself

Code is for machines. Docs are for humans.

Your team writes code for today, but people read documentation for years.

A good technical writer turns tangled Slack threads, undocumented design decisions, and tribal knowledge into clean, searchable documentation that the whole team can use — like architecture overviews, integration guides, or decision logs.

4- They Bridge the Gap Between Engineers and Non-Engineers

Product managers, support staff, QA testers, even clients — they all interact with your system in some way. But they don’t read code.

Technical writers create shared language between technical and non-technical people. That alignment helps projects run smoother, feedback loops shorten, and handoffs happen without confusion.

5- They Make Technical Work Last

Code doesn’t last forever. People leave. Rewrites happen. Teams evolve.

What stays behind is the reasoning — the why behind the build. Unless it’s undocumented.

A technical writer makes sure your systems are not just built well, but remembered well. That means fewer regressions, faster reworks, and better long-term decisions.

"But We’re a Small Team…"

I hear this all the time — and I get it. You’re shipping fast, wearing multiple hats, and trying to stay lean.

But that’s exactly when documentation matters most.
Small teams don’t have time for miscommunication. They need to move quickly with clarity, not confusion. And they can’t afford to keep answering the same questions every week.

Even a part-time or embedded technical writer can make a huge difference.

Documentation Is Infrastructure

Whether you’re scaling a startup or managing a mature system, clear documentation isn’t a nice-to-have — it’s infrastructure.

A good technical writer doesn’t slow your team down. They speed it up.
They preserve your hard-earned knowledge.
They make your systems understandable, usable, and maintainable — by everyone who needs to touch them.

This is something I see all the time in my work — I’ve helped engineering teams go from “Where do we even start?” to having clear, usable documentation that supports smoother handoffs, fewer questions, and more confident devs.

If you're thinking about how to make your systems more scalable — consider starting with how you explain them.

✍️ Over to You
Do you work with a technical writer?
Are you the de facto writer on your team?
Or is documentation something that keeps getting pushed down the priority list?
Let me know — I’d love to hear how your team handles it.

Also if you or your team needs a technical writer, please contact me.

How to build a Website on Wordpress.com (Step-by Step Guide for Beginners)

SandraMeshack — Tue, 02 Sep 2025 15:42:54 +0000

Introduction
Building a professional website doesn’t have to be overwhelming. With WordPress.com, you can launch a fully functional, mobile-friendly website without writing a single line of code. Whether you’re a small business owner, freelancer, or blogger, this guide will walk you through the entire process — from choosing a domain to optimizing your site for SEO.

By the end of this guide, you’ll have:
✔ A fully set up WordPress.com site
✔ A clean, responsive theme that reflects your brand
✔ Basic SEO and performance optimizations
✔ A launch checklist to make sure everything’s ready to go live

Let’s get started.

Step 1: Pick Your Domain Name

Enter your desired domain name in the search bar.
Choose "sitename.com" or another relevant extension or.
Connect your custom domain(domain you already own) or use a free yourname.wordpress.com for now.

Pro Tip: Keep it short, easy to spell, and brand-aligned.

Step 2:Sign up and Choose a Plan

Go to Wordpress.com.
Click "Get Started" and create your account using an email address or Google login.
Choose a plan:

Free: Good for personal blogs. Limited features and ads.
Personal or Premium: Custom domain. no ads.
Business or eCommerce: For custom themes, plugins, advanced analytics.

Pro Tip: If you are serious about branding, start with Premium or Business and custom domains. I just finished building a website for a client with premium plan.

Step 3: Select a Theme

Navigate to Appearance -> Themes in the Wordpress dashboard.
Filter by Business, Portfolio, or Blog themes depending on your needs.
Click Activate once you find the one that fits.

Step 4: Set Your Site Title, Tagline and Site Icon

Go to Settings -> General.
Enter your Site Title (your brand name) and Tagline (a short description).
Upload your Site Icon.

Step 5: Customise Branding (Colours, Fonts)

Navigate to Appearance -> Customise.
Upload your Logo, select brand colours, and adjust for readability.
Use site editor as this makes it easier for you to have a live update of your edits

Pro Tip: Stick to two fonts and two primary colors for a professional look.

Step 6: Create Core Pages

Go to Pages → Add New and create:

Home
About
Services / Products
Contact
Blog (optional but great for SEO) N.B. You can add as many pages as you require.

Step 7: Build the Homepage

Use WordPress Block Editor (Gutenberg)or Theme site editor to add:

Hero Section: Headline + Call-to-Action
Services Section: What you offer
Testimonials / Social Proof
Contact Button

Pro Tip: Use Columns, Image Blocks, and Buttons for a clean layout.

Step 8: Set Navigation Menu

Go to Appearance -> Menus.
Add your core pages and arrange them in order.
Enable Primary Menu and Footer Menu.

Step 9: Add Contact Form

Install the built-in Contact Form Block.
Link it to your email for inquiries.

Step 10: Configure Site Settings (SEO & URLs)

Go to Settings → Reading.
Set Homepage to your custom home page.
Configure permalinks for clean URLs: yoursite.com/page-name.

Step 11: Basic SEO Setup

Add SEO Titles & Meta Descriptions (in Settings → Site Info or using WordPress.com SEO tools(I used Yoast SEO).
Use H1 for main titles and H2 for subheadings.
Add alt text to all images.

Step 12: Optimise for Mobile

Preview your site in mobile view.
Adjust blocks and text for responsiveness.

Step 13: Add Analytics & Tracking

For basic stats, enable WordPress.com Stats.
For deeper insights, connect Google Analytics (Business plan required).

Step 14: Test Performance

Use PageSpeed Insights or GTmetrix to check load time.
Compress images and avoid heavy plugins.

Step 15: Launch Your Site

Preview your site one last time.
Make sure privacy settings are set to Public.

Hit Publish — your site is live!

✅ Final Launch Checklist

Domain connected
Theme customized
Pages created (Home, About, Contact, Services)
Menu set
Contact form working
SEO basics (titles, meta, alt text)
Mobile-friendly
Analytics installed
Speed tested
Privacy set to Public
Include Jetpack CRM if you want a CRM.

Please note that most of the images used for this tutorial were from a client's CRM website I developed using wordpress.com. View the Website here.

Next Steps

Want a professional, fully customised website without the stress?
👉 visit my website and contact me for WordPress design, backend scalability, and technical documentation services.

Comment if you have questions and I'd be more than happy to help.

Disclaimer:
This guide was created independently and is intended for educational purposes only. WordPress and WordPress.com are registered trademarks of Automattic Inc. I am not affiliated with, endorsed by, or in any way officially connected to Automattic or WordPress.com. All information provided here is based on personal experience and publicly available resources.

[Boost]

SandraMeshack — Fri, 15 Aug 2025 10:27:08 +0000

SandraMeshack

Jul 24 '25

🚀 Migrating Gmail to Gmail Using Google Takeout + Thunderbird (Mac-Friendly)

#thunderbird #emailmigration #google #devlive

Comments 1

3 min read

🚀 Migrating Gmail to Gmail Using Google Takeout + Thunderbird (Mac-Friendly)

SandraMeshack — Thu, 24 Jul 2025 23:27:37 +0000

🧩 Why I Did This
I had to move all my emails from a university Gmail account to my personal Gmail before the uni shut down my access. I wanted everything: attachments, threads, labels — you name it and my uni blocked access to the pop function so I could not fetch my emails from the server. Hence, I had to find a different route.

I used a combo of:

📦 Google Takeout
🕊️ Mozilla Thunderbird
💻 A MacBook
💡 A lot of trial and error

✳️ TL;DR Tools You’ll Need

✅ Google Takeout
✅ Thunderbird Mail Client
✅ Free Gmail account or your existing personal Gmail account
✅ Mac Terminal (caffeinate was a lifesaver)
✅ A bit of patience

📦 Step 1: Use Google Takeout to Export Old Gmail

Go to Google Takeout.
Click “Deselect all” — then scroll down and select only “Mail”.
Optional: You can click the “All Mail data included” button to pick specific labels.
Scroll down and click Next Step.
Choose:
Delivery method: Send download link via email
File type: .zip
File size: Set max to 2GB (⚠️ Thunderbird chokes on larger .mbox files)
Click Create export and wait. It might take a few hours depending on email volume. Google will email you a .zip file.

🗂️ Step 2: Unzip and Prepare MBOX File

Once downloaded:

Unzip the archive
You’ll find a folder like: Takeout > Mail > All mail including Spam and Trash.mbox
Optional: Rename that .mbox file to something simple like: old-emails.mbox

🕊️ Step 3: Install Thunderbird and Set Up Import

Download and install Thunderbird
Open Thunderbird and create a new email account (login to your new Gmail or whatever account you are moving the emails to)
- Use Gmail credentials (OAuth)
- Allow Thunderbird access via your browser when prompted
Install ImportExportTools NG
- Go to ☰ (hamburger menu) > Add-ons and Themes
- Search for: ImportExportTools NG
- Install and restart Thunderbird

📥 Step 4: Import MBOX to Thunderbird

In the left panel, right-click “Local Folders”
- choose ImportExportTools NG
- choose Import MBOX file
- then “Individual mbox files”
Select your "old-emails.mbox" file
It should create a folder under Local Folders with your imported mail.
⚠️ Issue I faced: Thunderbird said “Import Done 0/1” or showed “No messages found.”
✅ Fix: I had to unzip the file manually and make sure the .mbox file was not inside a nested folder. Then I reimported. It worked.

✈️ Step 5: Copy Emails to Your New Gmail Inbox

Once the MBOX shows messages under Local Folders, you can drag and drop them into your Gmail account folders within Thunderbird. Example: Drag → Local Folders > old-emails Drop → yourname@gmail.com > Inbox (or create a new folder)

💤 Pro tip: If you’re copying a lot, use Terminal’s caffeinate command to keep your Mac awake:

Error	Fix
MBOX import says 0/1	Rename MBOX file, make sure it's not zipped or inside subfolders
Thunderbird times out	Enable `caffeinate`, avoid sleep mode
Nothing appears after copy	Give it time — large volumes take a while to sync
Can't find “Import MBOX”	Must install `ImportExportTools NG` first
.mbox is >2GB	Go back to Takeout, split into smaller zip files (under 2GB)

✅ Final Tips

Keep Gmail open in a browser to monitor sync progress.
Thunderbird’s copy-paste to Gmail can take hours if you’re moving thousands of messages. Be patient — it works.
Once migrated, you can label, archive, or search your emails from Gmail as usual.

🎉 Done!

You’ve successfully migrated your email like a pro — no scripts, no third-party sketchy tools, just solid open-source software and smart file handling.

📌 Want to Reuse This?

Feel free to fork or share this on your blog or dev space. You earned it. And your inbox is lighter — metaphorically.

How I Help Engineering Teams Move Faster (and Breathe Easier)

SandraMeshack — Thu, 03 Jul 2025 23:24:12 +0000

Hi! I’m Sandra Chime — a software engineer and technical writer who’s passionate about making complex systems easier to use, adopt, and scale.

For the past few years, I’ve been the person on engineering teams who asks:

"How will the next developer understand this?"

"Can we make this API less of a mystery?"

"What happens when the person who built this leaves?"

If your team is scaling fast, maintaining a legacy codebase, or onboarding new devs, there’s a good chance your documentation is lagging behind. I help fill that gap.

✨ What I Bring to the Table

🧠 Technical Understanding

I’ve built and maintained backend systems using Python, JavaScript, and PHP — so I don’t just write docs, I understand the code behind them.

📝 Documentation That Actually Gets Read

From API references to onboarding guides to internal platform docs — I write documentation that’s clear, structured, and actually helps people do their jobs faster.

💡 Empathy for Developers and Users

I love translating engineering complexity into something understandable, whether it’s for junior devs, cross-functional teams, or end users.

📚 Real Writing Experience

Check out my published docs, tutorials, and articles here. I've written about everything from neural networks to GitHub Pages.

🛠️ How I Can Help Your Team

Rewrite or create API documentation that’s clean, useful, and up to date
Develop or fix onboarding docs that help new engineers ramp up quickly
Turn internal tools and workflows into guides anyone can follow
Create developer-focused blog posts, tutorials, and whitepapers
Review or refactor backend-heavy code with fresh eyes

📬 Let’s Chat

I’m currently open to freelance or contract-based technical writing and engineering projects.

Whether you’re a startup founder, lead engineer, or part of a DevRel team — if you need someone who speaks both code and clarity, I’d love to help.

📩 Email me at email

🔗 Or explore my docs and writing at sandrameshack.github.io

Thanks for reading. If you’ve ever said “we really need to clean up our docs” — I’m here to help you do it.

— Sandra Chime

How I Built My Technical Writing Portfolio Website Using GitHub Pages

SandraMeshack — Mon, 21 Apr 2025 02:06:24 +0000

How I Built My Technical Writing Portfolio Website Using GitHub Pages

As a technical writer and software engineer, I wanted a portfolio site that showcased my skills, articles, and publications in a way that felt modern and personal. I built it with HTML + CSS, hosted it on GitHub Pages, and added extra features like animations, Google Search Console, Analytics, and a sitemap.

Here’s exactly how I built it, step-by-step — no frameworks, no CMS, just code and care.

🛠️ Tools I Used

HTML + CSS — fully custom, no frameworks
GitHub Pages — free, fast hosting
Google Search Console — to help get discovered on Google
Google Analytics (GA4) — to track visitors
VS Code — for writing code
A simple animated avatar PNG — for personality ✨

🧱 Step 1: Create a GitHub Pages Repo

Go to https://github.com
Create a new repo named:yourusername.github.io
Make sure it’s public.
You can either:
Clone it locally (git clone)
Or drag-and-drop files directly into GitHub’s interface

💻 Step 2: Create `index.html`

This is the homepage. Here's the full code I used (copy it to a file called index.html):

<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
<title>Your Name | Technical Writer</title>
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;600&display=swap" rel="stylesheet">
<link rel="stylesheet" href="style.css">

<!-- Google Analytics (replace with your ID) -->
<script async src="https://www.googletagmanager.com/gtag/js?id=G-XXXXXXXXXX"></script>
<script>
 window.dataLayer = window.dataLayer || [];
 function gtag(){dataLayer.push(arguments);}
 gtag('js', new Date());
 gtag('config', 'G-XXXXXXXXXX');
</script>

<!-- Google Search Console verification -->
<meta name="google-site-verification" content="your-verification-code" />

<!-- Smooth scroll, link animation, fade effect -->
<style>
 html {
   scroll-behavior: smooth;
 }
 @keyframes fadeSlideUp {
   from { opacity: 0; transform: translateY(20px); }
   to { opacity: 1; transform: translateY(0); }
 }
 #work-section {
   animation: fadeSlideUp 1.5s ease-in-out both;
 }
 h2 {
   position: relative;
   animation: slideIn 0.9s ease forwards;
   opacity: 0;
 }
 @keyframes slideIn {
   from { opacity: 0; transform: translateX(-20px); }
   to { opacity: 1; transform: translateX(0); }
 }
 a {
   color: #4dd0e1;
   text-decoration: none;
   position: relative;
   transition: color 0.3s ease;
 }
 a:hover {
   color: #ffffff;
 }
 a::after {
   content: "";
   position: absolute;
   width: 0%;
   height: 2px;
   left: 0;
   bottom: -4px;
   background: linear-gradient(to right, #4dd0e1, #f97316);
   transition: width 0.4s ease;
 }
 a:hover::after {
   width: 100%;
 }
 .emoji-bg {
   position: absolute;
   top: 120px;
   right: 30px;
   font-size: 3rem;
   opacity: 0.05;
   z-index: 0;
   pointer-events: none;
   animation: float 6s ease-in-out infinite;
 }
 @keyframes float {
   0% { transform: translateY(0px); }
   50% { transform: translateY(-15px); }
   100% { transform: translateY(0px); }
 }
</style>
</head>

<body>
<div class="emoji-bg">🧠</div>

<nav>
 <div class="logo-container">
   <img src="avatar.png" alt="Logo" style="width: 60px; height: 60px; border-radius: 50%;">
   <strong>Your Name</strong>
 </div>
 <div>
   <a href="index.html">Home</a>
   <a href="contact.html">Contact</a>
   <a href="https://github.com/yourusername" target="_blank">GitHub</a>
 </div>
</nav>

<main>
 <h1>Hello, I'm [Your Name]</h1>
 <p>
   I’m a software engineer and technical writer with a focus on backend architecture and developer experience.
   I build scalable systems and write structured documentation to make complex tools easier to use.
 </p>

 <h2>What I Do</h2>
 <ul>
   <li><strong>Technical Writing:</strong> API docs, onboarding guides, tutorials</li>
   <li><strong>Software Engineering:</strong> Python, JavaScript, and backend systems</li>
   <li><strong>Dev Tools:</strong> Markdown, LaTeX, Confluence, GitHub Actions</li>
 </ul>

 <h2>Featured Articles</h2>
 <ul>
   <li><a href="https://dev.to/yourusername/article-title">How I Built This Portfolio</a></li>
   <li><a href="https://dev.to/yourusername/api-doc-template">API Documentation Template</a></li>
 </ul>

 <h2>Publication</h2>
 <p>
   <a href="https://example.com/publication-link">The Influence of Micro-expressions on Deception Detection</a>
 </p>

 <h2>Why I Do This</h2>
 <p>
   I’m driven by a passion for making technology accessible. My goal is to bridge the gap between complex systems and the people who use them — whether they're developers or stakeholders.
 </p>

 <section id="work-section">
   <h2>Let's Work Together</h2>
   <p>
     I’m available for freelance and contract-based technical writing. Email me at  
     <a href="mailto:youremail@example.com">youremail@example.com</a> or check out  
     <a href="https://github.com/yourusername" target="_blank">my GitHub</a>.
   </p>
 </section>
</main>

<footer>&copy; 2025 Your Name</footer>
</body>
</html>

🎨 Step 3: Add Styling in style.css

copy this to a file called style.css

body {
  background-color: #121212;
  color: #f5f5f5;
  font-family: 'Inter', sans-serif;
  margin: 0;
  padding: 0;
}

nav {
  background: linear-gradient(to right, #6a11cb, #2575fc);
  padding: 1rem 2rem;
  display: flex;
  justify-content: space-between;
  align-items: center;
}

📬 Step 4: Add a Contact Page

Create a contact.html file and mirror the design — just include your email address and links.

🌍 Step 5: Add Your Sitemap

Create a file called sitemap.xml:

<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
  <url>
    <loc>https://yourusername.github.io/</loc>
  </url>
  <url>
    <loc>https://yourusername.github.io/contact.html</loc>
  </url>
</urlset>

Upload it to your GitHub repo root.

📈 Step 6: Submit to Google

Go to Google Search Console
Add your site as a URL prefix
Choose HTML tag → Paste into
Go to “Sitemaps” → Add sitemap.xml
Done!!!.

Let me know if you want to build something similar! I’d love to help others make docs look as great as the code they explain.

Scientific Writing- Latex or Latex?

SandraMeshack — Fri, 16 Feb 2024 10:10:36 +0000

I am certain the first question that comes to mind is, 'What does Sandra mean by scientific writing and what has that got to do with tech or software development?'. Well, it gives me joy to say that there are developers and technical people right now who are also amazing researchers and scientists- I mean what are the odds?. Enjoying both worlds!!!.

So Scientific writing like technical writing is a form of writing that can be used to relay information to other scientists or technical people. The aim of this write-up is not to bash any system or to speak little of a system, rather, it's to introduce you to a better way of doing things.

Have you ever been handed a paper or document and by the looks of the formatting, you just assume there will be little to no reliable information in that document?- just because of how tacky it looks. Yes i admit, that is a bias but it happens and I would not want to dedicate time into writing something profound just for it to be shoved aside because it was not properly formatted or it does not look professional.

This is the point where i introduce you to Latex!!!. Latex is a high quality typesetting system that includes features designed for the production of scientific and technical documentation. It is free to use and you can learn how to use it here.

However, for people who love collaboration and version control, there is an online latex platform called overleaf. You can also learn how to use overleaf and latexhere.

A task for today, attempt writing a scientific or technical document with overleaf. When you get stuck, google is your friend. Do the tutorials!!!

Come back here to tell me what you think. Feel free to leave questions- I am highly versed with Latex and Overleaf is my friend. lol.

Have a happy professional writing and may your documents look as highly skilled as you are. Cheers!!!

Sample Code on how to build a ten neuron SNN USING STDP Learning rule

SandraMeshack — Wed, 19 Apr 2023 13:10:27 +0000

import numpy as np
import matplotlib.pyplot as plt

# Parameters
num_neurons = 10
num_layers = 3
num_timesteps = 100
dt = 0.001

# Neuron model
V_rest = -70
V_threshold = -55
V_reset = -75
tau_m = 0.02
tau_ref = 0.002

# Synaptic connections
W_ff = np.random.rand(num_neurons, num_neurons) * 0.1
W_fb = np.random.rand(num_neurons, num_neurons) * 0.1

# Spike-Timing Dependent Plasticity (STDP) learning rule
tau_stdp = 0.02
A_plus = 0.01
A_minus = -0.01

# Input data
X = np.random.rand(num_neurons, num_timesteps)

# Initialize variables
V = np.zeros((num_neurons, num_timesteps))
V[:, 0] = V_rest
spikes = np.zeros((num_neurons, num_timesteps))

# Simulation loop
for t in range(1, num_timesteps):
    # Calculate feedforward current
    I_ff = np.dot(W_ff, spikes[:, t-1])

    # Calculate feedbackward current
    I_fb = np.dot(W_fb, spikes[:, t-1])

    # Calculate total input current
    I_in = X[:, t] + I_ff + I_fb

    # Update membrane potential
    V[:, t] = V[:, t-1] + (-V[:, t-1] + V_rest + I_in * dt / tau_m) / tau_m

    # Generate spikes
    spike_indices = np.where(V[:, t] > V_threshold)[0]
    spikes[spike_indices, t] = 1
    V[spike_indices, t] = V_reset

    # Refractory period
    refractory_indices = np.where(spikes[:, t-1] == 1)[0]
    V[refractory_indices, t] = V_rest

    # STDP learning rule
    for i in range(num_neurons):
        for j in range(num_neurons):
            if spikes[i, t-1] == 1 and spikes[j, t] == 1:
                delta_w = A_plus * np.exp(-(t - 1 - tau_stdp) / tau_stdp)
                W_ff[i, j] += delta_w
            elif spikes[i, t-1] == 0 and spikes[j, t] == 1:
                delta_w = A_minus * np.exp(-(t - 1 - tau_stdp) / tau_stdp)
                W_fb[j, i] += delta_w

# Plot results
plt.figure(figsize=(10, 6))
plt.subplot(2, 1, 1)
plt.title("Membrane Potential")
plt.plot(V.T)
plt.xlabel("Time")
plt.ylabel("Voltage (mV)")
plt.ylim(V_reset - 10, V_threshold + 10)
plt.subplot(2, 1, 2)
plt.title("Spikes")
plt.plot(spikes.T, ".")
plt.xlabel("Time")
plt.ylabel("Spike")
plt.show()

Please note that you can always change the parameters based on the type of network you are building and you can always change the learning rules too. Please run this, make changes that suit your network and let me know what you have done.
Happy coding.

Machine Learning for Beginners

SandraMeshack — Thu, 16 Jun 2022 11:48:19 +0000

It's amazing how in the tech world and the non-tech world, everywhere you turn, people say AI(Artificial Intelligence), ML(Machine Learning) etc. I know for some, it's more or less thought to be buzz words while others have really gotten interested in knowing what this is all about and what they can do with Machine Learning. The question is, is ML as cool as people(techies) make it sound? YES it is!!!! Is there a barrier to entry? NO there is none!!!. There is room for basically everyone!

Main Question, How do i learn how to code and design ML algorithms?

The first thing, i'd advice is learn the basics or just learn the history. I have written an article on the history of neural network.
Learn python!!! Python is the most used language for ML in the industry and if you want to go into machine learning, you should learn python. There are numerous sites you could learn from and there are numerous youtube videos. However, one i've always given to my students and they have come back better is the learn python.org.
Practice Practice Practice!!!: Learning to code is more or less learning to use your hands at all times. Practicing and having pet projects will make you the better.
Google is your best friend: as a techie, you should learn to research. Learn to look at other people's codes, learn to look for solutions when you get stuck- getting stuck is a good thing and you should learn to ask questions.
There are numerous courses on learning ML. Use number 4 above to find one you like. One interesting course i always ask my students or new techies to use is introduction to tensorflow for deep learning.
As a learner forget about downloading an IDE(integrated development environment) for now. Also, use number 4 to figure out what an IDE is. What i'd suggest you use for learning is either Jupyter lab or google Colab. However, i strongly recommend google colab.
All you need is a good internet and your laptop for this.
PET PROJECTS!!!!- have a pet project!. There should be something you are working. Practice really does make perfect.

On the next blog post, i'd be sharing one of my pet projects and also sharing my thought processes with you.

Goodluck!!! Looking forward to seeing your projects!

instal drupal and xampp on ubuntu to enable the running of a full stack application on amazon web services

SandraMeshack — Thu, 03 Mar 2022 02:25:31 +0000

Starting an Instance
The first thing is to start an amazon instance and this can be achieved by following the steps listed below:
•Go tohttps://aws.amazon.com/console.
•Log into your aws account withcredentials(username/password).
•Go to services and click on instances.
•Create a new instance and give it a name of your choosing.
•Choose ubuntu as the operating system for web browser and launch ubuntu.
•Create a new key pair and give it a name of your choice.•
Download the key pair (This can only be downloaded once)

Firewall Security settings
After creating the instance and downloading the instance, the firewall security settings would need to be edited. Thiscan be achieved by following the steps below:
•Click on the running instance by selecting the instance box.
•Scroll towards the left side of the page until security groups is seen.
•Click on the security group of the instance.•Click on inbound at the bottom of the screen.
•Add http rule for web server and change the source to anywhere.
•Add https rule for web server and change the source to anywhere.
•Add SSH rule for web server and change the source to anywhere.
•Add Custom/TCP rule for web server, give it port 8080 and change the source to anywhere.

Linking instance to Elastic IP
In order to get a stable IP, link the amazon instance with an elastic IP with the steps listed below.
•Scroll down the left side of the page until you see network and security.
•Click on Elastic IPs.
•Click on Allocate new address.
•Click on Associate new address.
•Choose the instance you created earlier.

Connecting to Ubuntu
This document focuses on what was done using a windows computer therefore connecting to ubuntu and the procedure may be different for other operating systems.To achieve this follow the underlisted steps:
•Go to the folder where the downloaded key-pair was stored.
•Hold down the shift button of your computer and right click on the folder.
•Click on open with windows power shell.
•ssh -i "key pair" ubuntu@"elastic ip".
•Sudo su(connecting to the root server)
•Update Ubuntu Os Package by running this command - "sudo apt-get update".

Download Xampp for Ubuntu
64 bits Xampp for Ubuntu was downloaded and used for this project. This was achieved by following the description below.
•Go to https://www.apachefriends.org/download.html.
•Right click on xampp for linux and copy link address.
•Use the ’wget’ function to download xampp directly into ubuntu by running the following command ’wget "copied link"’.

Executing Xampp Connection
Xampp connections can be executed by following the below instructions.
•Run this command "sudo ./"file copied above".
•say yes to all the notifications.
•Run Xampp by executing: sudo /opt/lampp/lampp start
•Go to http://elasticIpaddressgottenfromaws.
•Notice that there is an access forbidden screen because xampp hasn’t been configured.

Configuring Xampp
Xampp can now be configured by running the following commands.
•vi /opt/lampp/etc/extra/httpd-xampp.conf
•Scroll to the end of the page and follow the instructions.

Restarting Xampp and Updating Xampp Security Settings
Restart Xampp to enable the latest changes take effect.
To restart Xampp and update security settings,run the followingcommands.
•sudo /opt/lampp/lampp restart .
•sudo /opt/lampp/xampp security .
•Follow the command.

Introduction to Spiking Neural Network

SandraMeshack — Fri, 25 Feb 2022 15:30:49 +0000

Spiking neural networks have been broadly described as the third generation of neural networks and the more computationally advanced generation of neural networks as opposed to the first and second generation. The mathematical model for spiking neurons could be traced back to over a hundred years ago when a paper on the excitability of nerves was published by Louis Lapicque. Spiking Neural Networks(SNN) is a special and enhanced category of Artificial Neural Networks where the communication and interaction of the neuronal models is done via series of spikes. The architecture of SNN is usually similar to the architecture of ANN but is largely inspired by biological properties hence has the ability to process computations in a more biological way. Studies have shown that spiking neuron networks possess good temporal data processing capability and are useful in the study of the pattern and behavior of the biological neural networks and provide useful tools for analysing elementary processes in the brain such as information processing, plasticity and learning. The focus of Spiking neural network is the development of a more biological plausible neuronal model. A phenomenological model such as spike response model is used in modelling the processing unit of SNN which is a spiking neuron while the transfer of information between neurons is done by electrical impulses known as spikes or action potentials that are released when the neurons are excited. Since trains of spikes are used in SNNs, they are capable of handling large amount of data while capturing the dynamics of the biological neurons.
The representation and the integration of several information like the time dimension, frequency and phase is also done by the SNN and they also provide solutions in applied engineering such as classification, event detection, signal processing, spatial navigation, speech recognition, data analytic, fault-tolerant computing and robotic control. The complex computational ability of the biological brain is not totally represented by the mathematical models for spiking neurons; however they are more realistic than the first and second generation of the neural network models because they tend to depict the output of the biological neuron which aids in theoretically investigating the use of time as resource for both interaction and computation.

Feed Forward Networks: The interest into the behaviour of feed-forward network had a dramatic increase after the backpropagation learning algorithm was discovered. In feed forward networks, there is no cycle formed in the connections to the nodes. The information in the feed forward network is uni-directional as information moves from one layer to the next forward layer.
Recurrent Networks: Recurrent networks have been used for the investigation of neural information processing that has to do with the formation of the associative memories which are also called working memory. Research has shown that recurrent networks could be difficult to develop and train although they are considered when dealing with models that have to do with decision making or sequence prediction problems.
Hybrid Networks: Hybrid network consists of a network with both feed forward and recurrent topologies. The interactions between the sub-populations could either be one directional or reciprocal. The types of hybrid networks shall be discussed in further write-ups.

The topology of the spiking neural network in the brain dynamically changes due to the learning process. This write up only deals with the introduction, history and topology of spiking networks while the review of models using spiking neural network shall be discussed in subsequent write-ups. Spiking Neural Network is more computationally advanced as earlier stated and has the ability to process computations better as it is largely inspired by the human mind. However, the SNN lacks the ability to self-repair like the human mind. Hence, researchers have gone ahead to introduce a new type of SNN called the Spiking astrocyte neural network which shall be introduced in the next write-up.

REFERENCES

A. Kasinski, “Introduction to spiking neural networks: Information processing, learning and applications,” Tech. Rep., 2011.
W. Maass, “Networks of Spiking Neurons: The Third Generation of Neural Network Models,” Tech. Rep. 9, 1997.
J. Liu, L. J. McDaid, J. Harkin, S. Karim, A. P. Johnson, D. M. Halliday, A. M. Tyrrell, J. Timmis, A. G. Millard, and J. Hilder, “Autonomous Learning Paradigm for Spiking Neural Networks,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11727 LNCS. Springer Verlag, 2019, pp. 737–744.
L. F. Abbott, “Lapicque’s introduction of the integrate-and-fire model neuron (1907),” Tech. Rep. 6, 1999.
N.BrunelandM.C.VanRossum,“Lapicque’s1907paper:Fromfrogs to integrate-and-fire,” Biological Cybernetics, vol. 97, no. 5-6, pp. 337– 339, 12 2007.
——, “Quantitative investigations of electrical nerve excitation treated as polarization,” Biological Cybernetics, vol. 97, no. 5-6, pp. 341–349, 12 2007.
S. Ghosh-Dastidar, H. Adeli, and A. G. Lichtenstein, “SPIKING NEURAL NETWORKS,” Tech. Rep. 4, 2009. [Online]. Available: www.worldscientific.com
A. P. Johnson, J. Liu, A. G. Millard, S. Karim, A. M. Tyrrell, J. Harkin, J. Timmis, L. J. McDaid, and D. M. Halliday, “Homeostatic fault tolerance in spiking neural networks: A dynamic hardware perspective,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 65, no. 2, pp. 687–699, 2 2018.
A. Taherkhani, A. Belatreche, Y. Li, G. Cosma, L. P. Maguire, and T. M. McGinnity, “A review of learning in biologically plausible spiking neural networks,” Neural Networks, vol. 122, pp. 253–272, 2 2020.
T. D. Sanger, “Optimal Unsupervised Learning in a Single-Layer Linear Feedforward Neural Network,” Tech. Rep., 1989.

DEV Community: SandraMeshack

What I Learned in My First Month of Owning a Digital Technology Company

Why Every Engineering Team Needs a Technical Writer (Even Small Ones)

How to build a Website on Wordpress.com (Step-by Step Guide for Beginners)

[Boost]

🚀 Migrating Gmail to Gmail Using Google Takeout + Thunderbird (Mac-Friendly)

🚀 Migrating Gmail to Gmail Using Google Takeout + Thunderbird (Mac-Friendly)

How I Help Engineering Teams Move Faster (and Breathe Easier)

✨ What I Bring to the Table

🛠️ How I Can Help Your Team

📬 Let’s Chat

How I Built My Technical Writing Portfolio Website Using GitHub Pages

How I Built My Technical Writing Portfolio Website Using GitHub Pages

🛠️ Tools I Used

🧱 Step 1: Create a GitHub Pages Repo

💻 Step 2: Create index.html

Scientific Writing- Latex or Latex?

Sample Code on how to build a ten neuron SNN USING STDP Learning rule

Machine Learning for Beginners

instal drupal and xampp on ubuntu to enable the running of a full stack application on amazon web services

Introduction to Spiking Neural Network

💻 Step 2: Create `index.html`