After eagerly anticipating the start of the program for ten days, I finally attended the first workshop on March 3rd and was thrilled to meet a diverse group of like-minded students from various universities in Copenhagen.

Throughout the program, we learned about the modern perspective of investors on startups and how to evaluate businesses based on factors such as team, market, business model, product, and traction. We also gained a deep understanding of complex terms like CHURN, CAC, LTV, MRR, ARR, and debt, and learned how to read financial roadmaps for each startup. One particularly intriguing topic was the valley of death, which we explored in depth.

We then delved into the entrepreneurship rollercoaster by analyzing the real experiences of Norlase. I realized that entrepreneurship is indeed a challenging journey that requires perseverance and resilience. However, it is also a rewarding experience that can bring about positive change in the world.

The final session of the program was presented by the co-founders of Lifeboard, who shared their incredible journey of saving lives and preventing drowning accidents. They started with no entrepreneurship experience, but through hard work and dedication, they were able to turn their prototype into a successful business.

Throughout the program, I learned that entrepreneurship is a continuous learning process.

No matter how much you know, there is always more to learn.

One question that particularly stuck with me was, “What gets you up in the morning?” It made me think deeply about my motivation for pursuing entrepreneurship and reminded me that a strong sense of purpose is essential to succeed in this field.

The program concluded with a networking session, which highlighted the importance of building a strong social network to succeed as an entrepreneur.

I left feeling inspired and motivated to continue my journey as a future entrepreneur.

In conclusion, this program provided me with valuable knowledge and skills and I look forward to continuing to learn and grow as an entrepreneur.

Stay tuned for the Ideation Workshop !

I was confident that I would be relocating to the United States to follow my aspirations after receiving acceptance letters from Carnegie Mellon University, Northeastern University, and Purdue University.

However, life is unpredictable.

I received two fully funded scholarships from European Union to pursue my Master’s across Europe.

There were many zoom calls and debates on where to go, and I made the right choice (Maybe)!

As I packed my bags and said goodbye to my friends and family, I couldn’t help but feel excited and nervous. I had no idea what to expect, but I was determined to make the most of this opportunity.

15th September 2022: Landed in a small town in Austria’s heart (Salzburg).

Welcome to the musical hub, whose rich history and love for music are celebrated throughout.

The first few days were difficult because it was pouring and cold (I went from a temperature of +40 to -4). I somehow gained autonomy walking to Ikea in the pouring rain with my classmates to get the essentials and grocery shopping. After two weeks, classes began, giving me adequate time to adjust.
I could critically evaluate the technology I love and its impact on society because of my coursework. My favorite “Digital Communication Theories” class addressed everything from the digital divide to filter bubbles. The class was anxious to learn and explore the myths of the digital world with guest lectures from the UK, Vienna, and Egypt.

3rd November 2022: RedBull IT Experience Day. I have been invited to attend the workshops and lectures in the Redbull Campus in Salzburg; I was the only non-german speaking person. However, the staff was very kind and ensured I had a great experience.

5th November 2022 : GastroHackathon at the Alles für den Gast in Salzburg. It was a great experience to brainstorm, design, and code for 24 hours. I learned IOS development for the first time and implemented a successful app. Damn, proud of myself.

Well, the first semester is almost finished. Grad school has been a wild rollercoaster, with many deadlines and proposals to write.

It has been one of my favorite Christmases ever. With international dinners, the autumn school in Wolfgangsee, visiting salt mines in Hallstatt, watching my first ice hockey match in the stadium, seeing snow for the first time, ice skating, visiting Christmas markets, and sipping mulled wine.

Wishing you a Happy New Year! May 2023 be as adventurous and thrilling!

Stay tuned for more!

The black box concept in ML is not new; sometimes, even we don’t know why the AI arrived at a specific decision.

So as the model gets advanced, it gets harder to explain how it works.

I recently came across this article that piqued my interest in asking, “if algorithms today are making the best decisions?”

In a typical AI system, there are a few steps :
First, training data is identified. Then, it undergoes the machine learning process. Then, a learned function or algorithm is produced. Through this algorithm, the system can make decisions or recommendations to a user for a specific situation.

But we may never know why did the AI do that? Why not something else? When did the AI succeed? When does it fail? When can I trust the AI? How can I correct an error? Well, it’s something to think about. No? Coming across this article illustrates how adaptive learning can be easily poisoned. So what’s the solution?

Before looking into the solution, let’s take a look into various incidents by AI:

Let’s take a look into different types of attacks and security risks of ML system:

1. Adversarial Training Data : It is also known as a Data Poisoning Attack where the attacker interferes with a model’s training data. ML models are feedback loops where new data is collected and then used to retrain the model. E.g., Microsoft experienced a very high-profile attack of this type in 2016 when it launched Tay, a Twitter chatbot aimed at teens.
2. Evasion Attacks : Also known as Perturbation Attack attempts to fool a model or cause it to misclassify an input. The attackers modify the input data to gain the desired response.
3. Membership Interference Attacks :This is the attack on the model’s API once it’s deployed and doesn’t require access to the training set/model weights. An attacker attempts to discover whether a particular person’s data was included in a model’s training set or simply whether any given training example was included in a training set.
4. Model Inversion Attacks :If the model has been trained on sensitive personal data that could cause harm if leaked, this poses a security risk and potentially breaches privacy.
5. Model Theft : Attacker seeks to recover the entire model through queries to its API.
6. Transfer learning Attacks: These freely available models are a potential security risk.

Now, let’s look into an overview to mitigate security risks:

1. General Security Practices.
2. Data Checks and Balances through validation, lineage tracking to record each source of data. Refer this.
3. Model Monitoring
4. Audits and Governance
5. Differential privacy
6. Encrypted ML

Well, no machine learning system can be guaranteed to be completely secure. So now what?

While reading more about the same, I came across the term XAI.

A formal definition: According to Wikipedia, Explainable AI refers to methods and techniques in the application of artificial intelligence technology such that the results of the solution can be understood by humans.

As can be seen in this diagram, an XAI system could replace the traditional learned function with an explainable model. This model would be built in a way where human beings could understand the decision-making process.

But, explanations can be hacked. Check this article.

So, what now?

Resources:

• The AI Transparency Paradox
• Is Building Secure ML Possible?
• Explainable AI: How humans can trust Artificial Intelligence
• What to Do When AI Fails

My iPhone uses Apple’s face tracking, and it is one of the AI conveniences making me wonder if my picture turns up somewhere on the Internet?

For any AI system, there are some essential dependencies. Let’s take a look.

Dependency 1: Data
Data is the lifeblood of AI. Without it, AI algorithms are useless.
Ever wondered what kind of data is used to train the systems that power today’s world ?

Well, let’s take a dive into them.

Dirty data refers to inaccurate, incomplete or inconsistent, especially in a computer system or database. Well, the year 2020 brought the recent evidence we see that in the case of AI systems that police us, dirty data and faulty algorithms can ruin lives, and our reliance on AI puts our liberty in great jeopardy.

Tech-washed racial bias followed by Faulty search data.

Your system is only as good as the data you use to train it.

Dependency 2: Algorithms

We’re all familiar with the term “algorithm.” Do we not? The author mentions Persuasive AI while discussing various algorithms that power advanced AI systems. I looked it up because I wasn’t sure what it meant.

It means that there is a science that studies what makes us want to do certain things, and when applied to predictive computer algorithms, it becomes persuasive AI. Persuasive technology is used in social media to predict and persuade us to take the most profitable action. All it needs to model us are the right algorithms and data. And, thanks to social media, we’ve willingly provided massive amounts of personal data that can be used to create our digital likeness.

Garbage in, garbage out is the first fundamental of computing and is a concept that every coder is familiar with. This book highlights that in AI, a discipline that is perceived to be highly complex, this fundamental rule is often forgotten. “While their algorithms and data are sure to produce imperfect results, when have we seen a human being ever make a perfect decision?” asks the author.

The author highlights an interesting case study: Sz Hua Huang v. Tesla Inc, resulting from Tesla’s self-driving feature, Autopilot. So the question arises If a human knows AI can be dangerous and uses it anyway, who is liable for the risk? Therefore, its correct to say that outcome is a product of intent. The intent of the software industry needs to be examined in order to produce a better outcome.

AI is critical, but it is also crucial to develop frameworks for responsible innovation.

So how can we innovate responsibly?

• Use AI only as needed.
• Require governance for AI
• Design Better User Interfaces
• Usability for Vulnerable Populations First
• Regulate AI bias

Bias that is allowed to remain in AI models will grow with the system. One widely publicized case of biased AI was Microsoft’s chatbot, Tay. Within 24 hours of Tay’s debut on Twitter, Tay learned to make racist, sexist, and lewd remarks from other users’ tweets. In response, Microsoft had to shut Tay down.

The fact is that just about anyone can build AI. We’d find it immensely alarming if anyone could easily gather the equipment, materials, and knowledge to build a nuclear bomb.

However, we don’t seem to be giving it much mind at all with AI. Why ?

As I sit today to write something,

I’m out of ideas.

Nothing.

Blank.

I’m asphyxiating in the social media feed; seeing my friends start a new adventure at new companies, cities, and even countries makes me question

where I am in life?

How can I measure my progress?

I feel I haven’t made any improvement because I focus on the results, the end goal, but I guess

I’m wrong here,

every day counts,
even the days I lie in bed,
even the days when I eat icecream,
even the days when I wake up early,
even the days when I watch the sunrise,
even the days when I hit the gym,
and today was one of those days.

The lyrics, “How can I swim when I’m drowning.” summarize the past few months of my life.

Sometimes one has to hit rock bottom to bounce back again.

So, hold on.

However, Kubernetes took the lead in defining how to run containerized workloads on a distributed amount of machines.

Kubernetes is widely adopted in the industry today, with most organizations using it in production. Kubernetes currently is a graduated CNCF project, which highlights its maturity and reported success from end-user companies.

What is Kubernetes?

Kubernetes is a portable, extensible, open-source platform for managing containerized workloads and services, that facilitates both declarative configuration and automation. It has a large, rapidly growing ecosystem.

Kubernetes provides :

Portability: It is a highly portable tool. This is due to its open-source nature and vendor agnosticism.

Scalability: Kubernetes has in-build resources, such as HPA (Horizontal Pod Autoscaler), to determine the required amount of replicas for a service. Elasticity is a core feature that is highly automated within Kubernetes.

Resilience: Kubernetes uses functionalities like ReplicaSet, readiness, and liveness probes to handle most of the container failures, enabling powerful self-healing capability.

Service discovery: Kubernetes provide cluster-level DNS (or Domain Name System), which simplifies the accessibility of workloads within the cluster.

Extensibility: Kubernetes is a highly extensible mechanism that uses the building-block principle.

The operational cost of the container: Operational cost refers to the efficiency of resource consumption within a Kubernetes cluster, such as CPU and memory. Kubernetes has a powerful scheduling mechanism that places an application on the node with sufficient resources to ensure the successful execution of the service. As a result, most of the available infrastructure resources are allocated on-demand.

Kubernetes is not a traditional, all-inclusive PaaS (Platform as a Service) system. Since Kubernetes operates at the container level rather than at the hardware level, it provides some generally applicable features common to PaaS offerings, such as deployment, scaling, load balancing, and lets users integrate their logging, monitoring, and alerting solutions.

Kubernetes architecture, composed of control and data planes.

A Kubernetes cluster is composed of a collection of distributed physical or virtual servers. These are called nodes.

Nodes are categorized into 2 main types: master and worker nodes.

The components installed on a node, determine the functionality of a node, and identifies it as a master or worker node.

The suite of master nodes, represents the control plane, while the collection of worker nodes constructs the data plane.

Control Plane

The control plane consists of components that make global decisions about the cluster. These components are the:

• kube-apiserver : the nucleus of the cluster that exposes the Kubernetes API, and handles and triggers any operations within the cluster
• kube-scheduler : the mechanism that places the new workloads on a node with sufficient satisfactory resource requirements
• kube-controller-manager : the component that handles controller processes. It ensures that the desired configuration is propagated to resources
• etcd : the key-value store, used for backs-up and keeping manifests for the entire cluster

There are two additional components on the control plane, they are kubelet and k-proxy. These two are special and important as they are installed on all node.

Data Plane

The data plane consists of the compute used to host workloads. The components installed on a worker node are the:

• kubelet : the agent that runs on every node and notifies the kube- apiserver that this node is part of the cluster
• kube-proxy : a network proxy that ensures the reachability and accessibility of workloads places on this specific node

Check out https://kubernetes.io/docs/concepts/overview/components/

After we write a code for an application, we host it into the server. Hosting involves running your web app on a particular server.
Hosting make the application available from a remote cloud infrastructure that is accessible through the Internet.

Now, How do we host applications?

In the past years, VMs (Virtual Machines) were the primary mechanism to host an application. VMs encapsulate the code, configuration files, and dependencies necessary to execute the application.
A virtual machine, commonly shortened to just VM, is no different from any other physical computer like laptops, smartphones, or servers. However, it has a CPU, memory, disks to store your files and can connect to the Internet.

Virtualization is not possible without the hypervisor. A hypervisor, or virtual machine monitor, is the software or firmware layer that enables multiple operating systems to run side-by-side, all with access to the same physical server resources. The hypervisor orchestrates and separates the available resources (computing power, memory, storage, etc.), aligning a portion to each virtual machine as needed.

However, using VM posed specific challenges as VMs require significant RAM and CPU resources.

As a result, the containers were introduced to make scaling up fast and easy.

Where shipping containers hold products comprising many different parts, a software container encapsulates an application — often a single executable service or microservice — along with its libraries, frameworks, and other components.

Multiple VMs on a hypervisor is replaced by numerous containers running on a single host operating system.

The processes in the containers are completely isolated but can access the OS filesystem, resources, and packages.

The creation and execution of containers are delegated to a container management tool, such as Docker.

Therefore, the containers are lightweight so that you can deploy multiple containers on a single server (or a VM) — no more dedicating an entire server to a single application. And you only have one OS to maintain. As a result, scaling up becomes fast and easy without the need for more server space.

What is a Container?

• A Self-contained, sealed unit of software.
• It contains everything required to run the code.
• It includes batteries and OS
• It has Code, Config, Processes, Networking, Dependencies, OS.

Consequently, there is a better usage of available infrastructure and a more efficient pathway to release a product to consumers.

The appearance of containers enabled organizations to ship products using a lightweight mechanism that would make the most available infrastructure. There are plenty of tools used to containerize services. However, Docker has set the industry standards for many years.

To containerize an application using Docker, three main components are distinguished:

• Docker files: A Docker file is a set of instructions used to create a Docker image.

• Docker images: A read-only template that enables the creation of a runnable instance of an application.

# build the image

docker build -t go-helloworld .

# tag the image

docker tag go-helloworld aucodes/go-helloworld:v1.0.0

# push the image

docker push aucodes/go-helloworld:v1.0.0

• Docker registries: The image needs to be pushed to a public Docker image registry to make it accessible to everyone.

Now, I assume you understood the basics, now onto the next phase.

The next stage in the release process is the deployment of the service. However, running an application in production signifies that thousands and millions of customers simultaneously consume the product. For this reason, it is paramount to build for scale. It is impracticable to manually manage thousands of containers, keeping these up-to-date with the latest code changes, in a healthy state, and accessible.

As a result, a container orchestrator framework is inevitable.

Checkout out the previous blog.

Stay Tune for more!

Resources :

https://docs.docker.com/develop/develop-images/dockerfile_best-practices/

https://docs.docker.com/engine/reference/builder/#from

https://docs.docker.com/engine/reference/commandline/tag/

https://buildpacks.io/docs/app-journey/

Before coding an application, it’s necessary to understand the design considerations.

It depends on :

• Business Requirements -> understanding who the stakeholders are, the functionalities, the input/output process, and finally, the end-users.
• Available resources -> which include engineering resources, financial resources, time frames, and internal knowledge.

So, before diving, let us take a look at the Application Architectural designs.

Monolithic architecture:

In the monolith, all the components are coupled to each other and distributed as a single unit. As a result, all the business logic functions have the same repository, share resources, and have the same programming language. As a result, the development is simple; deployment is straightforward, and communication between components happens in the same process.

But what happens when you have to scale the application?

Yes, you are correct; it becomes much more complicated to maintain it since the change in one component would imply modifying another. Likewise, the deployment becomes more complicated since a change means redeploying the entire system again. It’s no longer a simple task since, generally the only way to scale this type of application is to replicate the whole application over and over again.

Microservice Architecture:

This type of architecture proposes to have lightweight decoupled services with a single responsibility isolated from each other with independent storage and autonomous deployability.

According to Sam Newman, “Microservices are the small services that work together.”

The Application tiers include:

• User Interface -> Handles HTTP request.
• Business Logic -> refers to the functionalities for the specific service.
• Data Layer -> for communicating with the Database.

For better understanding, take a project, e.g., design an Airplane Booking System and compare the architectures.

There is no “golden path” to design a product, but a good understanding of the trade-offs will provide a clear project roadmap.

Let’s look into the trade-offs:

• Development complexity
• Scalability — focuses that there are enough replicas of an application to satisfy all user requests.
• Time to deploy — refers to ship new features efficiently using the pipeline.
• Flexibility
• Operational cost
• Reliability- ability to respond quickly to failures by having insightful logs and metrics.

So, what can be the best practices for Application deployment?

• Health Checks
• Metrics
• Logs
• Tracing
• Resource consumption

Always remember the choice of an application structure is highly impacted by available resources, requirements, and involved trade-offs.

Beginner ? Read the Introduction to Cloud Native.

wanna know more?

Resources :

https://microservices.io/

https://www.geeksforgeeks.org/monolithic-vs-microservices-architecture/

Becoming a software developer is not always about designing and coding applications; it’s also significant for any software developer to understand how the deployment process runs and how the infra operates.

Hence, understanding the fundamentals is of utmost importance.

As Cloud-native is among the most pervasive terms heard across IT today, let’s understand its basics.

What precisely is cloud-native?

Cloud-native is an approach that enables developers and organizations to be more agile, providing workload portability and scalability.

CNCF ( Cloud Native Computing Foundation ) defines it as Cloud-native technologies that empower organizations to build and run scalable applications in modern, dynamic environments such as public, private, and hybrid clouds. Containers, service meshes, micro services, immutable infrastructure, and declarative APIs exemplify this approach.

Four Pillars of Cloud-Native:

• Use micro service approach to build applications.
• Package micro service as containers.
• Use the Dev Ops approach to deploy infrastructure.
• Continuously deliver your applications to end customers.

Cloud-native relies on existing technologies and assumes that organizations have already undergone a paradigm shift to Dev Ops and continuous integration or delivery. The aim is to get the development team and operation teams to work better together.

Thus, operations get the ability to handle infrastructure in a much more agile manner. With features to manage resources automatically,quickly scaling the infrastructure and platform features to recover from application failures allows operations to be more resilient. This allows ops to spend more time building better infrastructure,managing machine utilization, and improving operational overhead in general.

Containers are closely associated with cloud-native terminology. Containers are used to run a single application with all required dependencies. The main characteristics of containers are easy to manage, deploy, and fast to recover. Kubernetes is a container orchestrator that is capable of solutionizing the integration of the following functionalities:

• Runtime
• Networking
• Storage
• Service Mesh
• Logs and metrics
• Tracing

The main reasons why an organization needs to adopt cloud-native technologies are enabling quick delivery of value to customers and quickly extending to new features and technologies.

There are two perspectives to keep in mind.

Business :

• Agility
• Growth
• Service availability

Technical perspective:

• Automation
• Orchestration
• Observability

Read Part 2.

Stay tuned for more!

How can we fight through the thoughts of the unpredictable future?

How can I tell my mind not to overthink every situation and question every choice?

With the onset of technology, every click, every dm, every snap somehow pushes me to compare myself with others, and the never-ending thoughts echo, “ I’m not good enough”.

Is it something we all go through?

Maybe.

I lie awake, bestowing my mind with insecurities, and the vulnerabilities and the uncertainty of the future terrify me.

“What if I never get what I want to achieve”

With continuous mindless scrolling through LinkedIn, Instagram; the anxious sensation returns with great strength, empowering me to overthink and cry.

A perpetual cycle.

What is the way out?

or

Is there a way out?

World, I want to leave you better
I want my life to matter
I am afraid I have no purpose here
I watch the news on TV
Abandon myself daily
I am afraid to let you see the real me

Rain it falls, rain it falls
Pouring on me
And the rain it falls, rain it falls
Sowing the seeds of love and hope, love and hope
We don’t have to stay here, stuck in the weeds

Have I the courage to change?
Have I the courage to change?
Have I the courage to change today?