Boost Your Data Science Skills: The Essential SQL Certifications You Need

Boost Your Data Science Skills: The Essential SQL Certifications You Need
Image by Editor

SQL, pronounced s-q-l or sequel, stands for Structured Query Language. It is a programming language used to access and manipulate databases. When working with large datasets, one of the issues can be extracting exactly what you need from it.

SQL helps with this. Professionals can communicate with relational database systems by cleaning, sorting, extracting and processing large datasets. The demand for SQL specialists is still rising due to the valued assets of big data and the need for professionals who can process, assess, and analyze it.

There are different types of SQL certifications which are platform-specific. Depending on what database technology you are interested in which helps you determine which certification best suits you.

Microsoft Certified: Azure Data Fundamentals

Link: Microsoft Azure Data Fundamentals

This course is aimed at beginners and can be completed in 1 month, committing 10 hours a week.

In this 5 course series, learn about the fundamentals of database concepts in a cloud environment. You will achieve basic skills in cloud data services, and build your foundational knowledge of cloud data services within Microsoft Azure.

The 5 courses are:

  • Explore Core Data Concepts in Microsoft Azure
  • Microsoft Azure SQL
  • Microsoft Azure Cosmos DB
  • Modern Data Warehouse Analytics in Microsoft Azure
  • Preparing for DP-900: Microsoft Azure Data Fundamentals Exam

Oracle SQL Databases

Link: Oracle SQL Databases Specialization

This course is aimed at beginners and can be completed in 1 month, committing 10 hours a week.

The Oracle SQL Databases specialization certification consists of 4 courses where you will learn how to create Oracle SQL Databases and understand how to insert, modify, and delete from Oracle SQL Databases.

The 4 courses are:

  • Oracle Database Foundations
  • Oracle Database Platform
  • Oracle SQL Basics
  • Oracle SQL Proficiency

IBM Data Warehouse Engineer

Link: IBM Data Warehouse Engineer

This course is aimed at beginners and can be completed in 3 months, committing 10 hours a week.

The IBM Data Warehouse Engineer certification consists of 8 courses where you will learn the most up-to-date practical skills and knowledge data warehouse engineers use, how to deploy, manage, secure, monitor and optimize relational database systems like MySQL and PostgreSQL. You will also gain knowledge of various types of SQL and queries to access and manipulate data in databases.

The 8 courses are:

  • Introduction to Data Engineering
  • Introduction to Relational Databases (RDBMS)
  • SQL: A Practical Introduction for Querying Databases
  • Hands-on Introduction to Linux Commands and Shell Scripting
  • Relational Database Administration (DBA)
  • ETL and Data Pipelines with Shell, Airflow and Kafka
  • Getting Started with Data Warehousing and BI Analytics
  • Data Warehousing Capstone Project

IBM Business Intelligence (BI) Analyst

Link: IBM Business Intelligence (BI) Analyst Professional Certificate

This course is aimed at beginners and can be completed in 4 months.

With the IBM Business Intelligence Analyst certification, you will be able to launch your Business Intelligence career in less than 4 months.

This certification is made up of 10 courses where you will learn about SQL queries, relational databases, data warehousing, data analysis as well as reporting techniques. You will learn how to apply statistical analysis methods to identify trends, create visualizations, and generate valuable insights, leading to informed decision-making and process improvement.

The 10 courses are:

  • Business Intelligence (BI) Essentials
  • Excel Basics for Data Analysis
  • Data Visualization and Dashboards with Excel and Cognos
  • Introduction to Relational Databases (RDBMS)
  • SQL: A Practical Introduction for Querying Databases
  • Getting Started with Data Warehousing and BI Analytics
  • Statistics Fundamentals Using Excel
  • Getting Started with Tableau
  • Advanced Data Visualization with Tableau
  • BI Analyst Capstone

PostgreSQL for Everybody

Link: PostgreSQL for Everybody Specialization

This course is aimed at the intermediate level and can be completed in 1 month, committing 10 hours a week.

This PostgreSQL for Everybody certification consists of 4 courses. In these courses, you will learn the basics to the advanced of SQL. You will dive into how to use the PostgreSQL database effectively, explore database design principles, database architecture and deployment strategies as well as be able to compare and contrast SQL and NoSQL database design approaches.

The 4 courses are:

  • Database Design and Basic SQL in PostgreSQL
  • Intermediate PostgreSQL
  • JSON and Natural Language Processing in PostgreSQL
  • Database Architecture, Scale, and NoSQL with Elasticsearch

Wrapping it up

Although knowledge and experience is prioritized over certification. It is important to know that hiring managers do prefer applicants who are certified. When choosing the right SQL certification, make sure you take into consideration the particular vendor or platform that the job you want it asking for.

Nisha Arya is a Data Scientist and Freelance Technical Writer. She is particularly interested in providing Data Science career advice or tutorials and theory based knowledge around Data Science. She also wishes to explore the different ways Artificial Intelligence is/can benefit the longevity of human life. A keen learner, seeking to broaden her tech knowledge and writing skills, whilst helping guide others.

More On This Topic

  • Data Science Minimum: 10 Essential Skills You Need to Know to Start…
  • Popular Certifications to validate your data and analytics skills
  • Boost Your AI and ML Skills for Free at NVIDIA Conference
  • In-demand SAS Certifications for Your Success
  • Data Analyst Skills You Need for Your Next Promotion
  • Want to Use Your Data Skills to Solve Global Problems? Here’s What…

How to avoid the headaches of AI skills development

brain made of yellow balloons

Adopting artificial intelligence (AI) to assist with tasks in business and technical functions looks good on quarterly reports for shareholders. But throwing technology into an organization doesn't deliver overnight miracles.

The big challenge with AI is that many business and IT leaders aren't sure whether their organizations are ready to handle it productively, according to a recent survey of 1,200 IT executives and professionals by Pluralsight.

Also: 5 ways to prepare for the impact of generative AI on the IT profession

While 81% of IT professionals are confident they can integrate AI into their roles, only 12% have significant experience working with the technology. To complicate matters further, 90% of executives don't completely understand their team's AI skills and proficiency.

"Even as organizations accelerate AI adoption, the majority don't understand what, if any, AI skills their employees possess or have an upskilling strategy to develop them," the study's authors point out.

"The AI skills gap doesn't only apply to advanced technical skills, either," the report continues. "To make the most of AI, organizations need an accurate way to benchmark AI skills across their organization and use their insights to create a plan for skill development. This should include basic AI literacy as well as hands-on experiences where employees can apply what they learn, experiment, and make mistakes in a safe environment."

Also: Want to work in AI? How to pivot your career in 5 steps
The AI skills gap impedes long-term success, the survey shows. Almost all executives and IT professionals (94%) believe AI initiatives will fail without staff who can use AI tools effectively.

Executives and IT professionals agree that investing in talent, training, and culture is the top step organizations should take to prepare for emerging AI tools. Yet only 40% of organizations have formal structured AI training and instruction.
Some leading organizations recognize the potential of AI and are taking proactive measures to address the challenge of preparing their workforces for AI — even employing AI itself to boost AI skills.

For example, Johnson & Johnson has launched a concerted effort to prepare employees for an AI-driven economy.

"We hire learners," says Jim Swanson, executive vice president and chief information officer at Johnson & Johnson (J&J).

Also: Have 10 hours? IBM will train you in AI fundamentals — for free

"Everyone in our organization has an objective that focuses on how they'll grow and develop their skills in 2024. Our global, AI-powered learning platform, J&J Learn, offers courses like a six-week digital immersion program to build capabilities in product management, design thinking, and artificial intelligence."

Core technology skills essential in today's AI era include software development, cloud engineering, data management, and network operations, says Swanson: "Just consider how foundational elements like data and elastic compute fuel the AI models that are currently in the spotlight."

Also: Six skills you need to become an AI prompt engineer

However, AI isn't just important for technology professionals. Swanson says everyone across the organization should play a role in digital growth.

"Leaders should take an active part in equipping their employees with critical future-ready skills, like how to responsibly apply generative AI to improve productivity, how to leverage intelligent automation to speed operations, or how to simulate steps in a supply chain with digital twins or augmented reality," he says.

J&J also incentivizes learning "through a month-long challenge where associates hone their technical and leadership skills, with points earned translating into donations for students in need globally," says Swanson.

"We believe that training is critical, but it is through experience that this upskilling takes its full dimension. We pair these digital upskilling courses with growth gigs and mentorships, providing the opportunity to reinforce learning through experience and exposure."

Also: Beyond programming: AI spawns a new generation of job roles

While AI is all about leading-edge technology, Swanson says the exploitation of tools is only part of the story. "Even as we reimagine how we work through technology, we will always need people in the center to guide our innovation," he says.

"The business of technology is distinctly human, and we need to foster soft skills like customer-centricity, people leadership, and communications."

The Pluralsight survey authors make the following recommendations to executives who want to build a well-trained technical and business workforce:

  • Consider upskilling versus outsourcing for AI capabilities — "Eighty percent of executives and 72% of IT practitioners agree their organization often invests in new technology without considering the training employees need to use it," the Pluralsight survey suggests. "Finding AI experts in the market is a gamble. Organizations that develop AI talent from their existing workforce will build the exact AI skills they need while providing valuable professional development opportunities to their teams."
  • Develop an AI training strategy before implementing AI applications — "Organizations that want to take advantage of emerging technologies don't always have time to train their employees first. But if they can implement an upskilling strategy before deploying AI technology, their teams can start driving value from day one."
  • Assess current AI capabilities — "Organizations need visibility into their teams' AI capabilities. Once they understand their strengths and weaknesses, they can develop an upskilling program that fills in the gaps and gives them the skills they need to use AI tools effectively."

Artificial Intelligence

Luminary Cloud’s simulator taps GPUs to help speed up product design

Luminary Cloud’s simulator taps GPUs to help speed up product design Kyle Wiggers 8 hours

Simulations are an essential step in physical product engineering. They enable engineers to create prototypes and understand how they might perform in the real world, accounting for factors like aerodynamic drag, air and water flows and pressure and temperature distributions.

But today’s engineering simulators tend to be quite slow and difficult to scale up, according to Jason Lango, the CEO of Luminary Cloud.

“Most engineers use legacy software that runs on on-premises infrastructure, resulting in a slow design workflow where each simulation takes days or weeks,” Lango told TechCrunch in an interview. “This decades-old legacy software hasn’t modernized to run on the cloud and GPUs.”

Luminary, a startup building a platform to run engineering simulations, by contrast does rely on the cloud and GPUs, Lango says — specifically clusters of Nvidia GPUs. Luminary’s simulation tooling, which is being used by customers including Puma golf brand Cobra Golf and Joby Aviation, the air taxi startup, lets engineers test engineering scenarios to optimize their product designs ostensibly faster than legacy software solutions.

“Luminary’s real-time engineering approach empowers engineering teams to perform product simulation and analysis cycles in minutes, rather than weeks,” Lango said. [This] results in improved time to market, faster insights, better design outcomes, increased team productivity and better use of valuable physical prototyping dollars.”

AI-leaning

Luminary isn’t exactly first-to-market with a cloud-based engineering simulation tool. Rivals include Siemens, Dassault Systèmes, PhysicsX, Simscale, Flexcompute and Ansys, which Synopsys recently acquired for $35 billion.

But one aspect that sets Luminary apart is its investment in AI, Lango says.

The platform offers an AI assistant, Lumi AI, that automatically handles tasks like mesh generation. Mesh generation — which divides an object or scene into small, discrete cells or elements called meshes — is a key component of the simulation process, explains Lango.

Luminary Cloud

Image Credits: Luminary Cloud

“Luminary is a simulation and analysis company, but we’ll eventually become a data company,” he continued. “With AI and machine learning capabilities, [our] platform could not only provide suggestions for how and where to make modifications to improve and optimize designs, but additionally learn how to set up, solve and visualize simulations better, allowing even novice users to extract significant insights.”

In another potential differentiator, Luminary doesn’t charge for a license or subscription fee — unlike some of its rivals. Instead, customers pay for simulation and analysis in dollars per minute of GPU usage.

“For example, a quick conceptual design simulation of an aircraft could run for a couple of minutes and cost less than $90,” Lango explains. “Customers doing larger, high-fidelity simulations, or large numbers of simulations exploring different design alternatives, would achieve volume for prepaid capacity discounts.”

Origin and expansion

Lango founded Luminary in 2019 with Juan Alonso, a Stanford aeronautics and astronautics professor. Prior to Luminary, Lango worked at Silicon Graphics, the high-performance computer chip manufacturer, as well as NetApp and Cisco, while Alonso ran NASA’s Fundamental Aeronautics Program, leading research in supersonic projects.

Lango met Alonso during summer 2019 through Sutter Hill Ventures, the private equity firm, where Lango was an entrepreneur in residence. Sutter Hill managing director Doug Mohr invited Alonso — at the time, one of his squash partners — to meet Lango at the firm’s offices in Palo Alto, suspecting that Alonso and Lango would make strong business partners.

Lango and Alonso clicked. And, over the course of several meetings, they refined the idea for the platform that became Luminary.

Sutter Hill sees a lucrative future in Luminary, evidently, as the startup’s customer base eclipses 33 organizations across the automotive, aerospace and defense and industrial equipment sectors.

Today, Luminary announced that it raised $15 million in equity and $100 million in debt from Sutter Hill, a tranche the company plans to put toward expanding its sales org and product capabilities.

“Jason and Juan are bringing the power of GPUs and the elasticity of the cloud to one of the most complex engineering functions,” Mike Speiser, managing director at Sutter Hill, said in an emailed statement. “Luminary Cloud has the potential to [overhaul] the expensive and arduous process of product development.”

Luminary, based in San Francisco, has a team of 73 staffers and aims to grow that number to 80 by the end of the year.

“Luminary’s on-demand and prepaid capacity consumption model is a great way to have a win-win with forecast volume and discounting,” Lango said. “Customers of any size can start with a $0 up-front on-demand relationship … [As a result,] generally, our sales haven’t been impacted by the slowdown in tech, also due to our customer base being industrial R&D and engineering companies continuing to invest in their own commercial or consumer product development.”

Google Unveils FAX for Scalable Federated Learning

The Real Reason Behind Google’s Recent Layoffs

Google Research has developed a new software library, FAX, that helps with big-scale computations for machine learning across different devices, such as computers and phones. It’s built on top of JAX, a tool for high-performance ML, and designed to facilitate working with federated learning.

Read the full paper here.

Why FAX Matters

.At its core, FAX is built upon JAX, a high-performance ML library. Leveraging JAX’s capabilities, FAX introduces several key features tailored to FL scenarios. This enables FAX to target both data center and cross-device applications, where computations may be distributed across various devices.

It provides tools to write and optimize ML computations easily. It uses techniques like automatic differentiation (AD) and just-in-time (JIT) compilation to make ML programs more efficient.

Additionally, FAX specifically targets federated learning, where devices collaborate on ML tasks without sharing data directly. This type of learning is important for privacy reasons and for handling large amounts of data.

Key Features

  • Scalability: FAX can handle large models and distribute computations across many devices, making it possible to handle massive models that may not fit within the memory constraints of a single device.
  • Easy Programming: It’s designed to make writing and optimising ML programs simple.
  • Automatic Differentiation: FAX includes tools for automatic differentiation, which is crucial for training ML models efficiently.
  • Production Integration: FAX can translate ML models into production systems that run on mobile devices.

While there are other frameworks for federated learning, FAX stands out because it combines scalability, federated AD, and integration with production systems in a single package. Other frameworks may focus on one or two of these aspects, but FAX aims to address all three simultaneously.

To sum up, FAX is a powerful tool for building and deploying ML models, particularly in scenarios where data privacy and scalability are important considerations.

The post Google Unveils FAX for Scalable Federated Learning appeared first on Analytics India Magazine.

AGI Coming Soon

AGI Coming Soon

Looks like AGI’s arrival is advanced by a year.

“We’re not quite there [yet], but we will be there, and by 2029 it [AI] will [be able to] match any person,” said Ray Kurzweil, the famous computer scientist, in a recent podcast with Joe Rogan, when asked about his views on AI and its capabilities. “I’m actually considered conservative. People think it might happen next year or the year after,” he added.

Kurzweil recalled his claim from 1999, when he had predicted that by 2029 AI would be able to match humans. “For 30 years, people thought that was totally crazy,” he added. “In fact, Stanford held a conference inviting several hundred people from around the world to talk about my prediction. People thought that this would happen, but not by 2029. They thought it would take 100 years.”

Elon Musk reposted on X agreeing with Kurzweil’s views. “By 2029 AI is probably smarter than all humans combined (sic),” he wrote. Lex Fridman reacted to this saying, “We are in for a few interesting years. I hope humanity wins in the end.”

AI will probably be smarter than any single human next year. By 2029, AI is probably smarter than all humans combined. https://t.co/RO3g2OCk9x

— Elon Musk (@elonmusk) March 13, 2024

Kurzweil further explained that people often underestimate the rate at which technology grows. “It actually doubles in fourteen years,” whereas, he said, people think it only grows by 2% every year. Talking about the speed of computers, he said that the calculation speed has increased to 35 billion calculations per second, which is a 24 quadrillion fold increase from 0.00007 calculations per second in 1939.

In August last year, Saturnin Pugnet, founding member of Tools for Humanity, the company behind Worldcoin, started the AGI conversation claiming that “Fusion and AGI are probably happening by 2030″.

Stating that unlimited energy and intelligence will come to fruition by the end of the decade, he instantly takes you back to what OpenAI CEO Sam Altman had predicted a couple of years ago. He said that the costs of intelligence and energy are going on a path towards near-zero and that by 2030 “AI revolution and renewable + nuclear energy will get us there”.

AGI has been achieved

The form of AGI has been the most-discussed topic online, with people giving their two cents about it.

The ongoing discussions has led to industry leaders weighing in their opinions as well. Elon Musk could not have chosen a better time to slap OpenAI with a lawsuit. Just as OpenAI and other tech titans invested in robotics company FigureAI, Musk filed a case against OpenAI and Altman, as it supposedly inches towards AGI.

The lawsuit claims that OpenAI has already achieved AGI with GPT-4.

Same was the case when OpenAI accidentally leaked Q* and everybody called it AGI. There is no doubt that OpenAI has been hung up on building AGI. Interestingly, the company took another step towards this goal and changed the “core values” listed on its website to add the focus on AGI, something that wasn’t explicitly mentioned on the page before.

Recently, NVIDIA head Jensen Huang shared his view on AGI as well, expressing his belief that its realisation is determined by how the goal is defined. “If I gave an AI every single test that you can possibly imagine, make the list of tests and put it in front of the computer science industry, I am guessing in five years’ time, we will do well on every single one,” he said.

Huang also emphasised that AGI could be farther away, and believes that it is hard to achieve it as an engineer, as engineers need defined goals. With the evaluation parameters still evolving, and with the moving definition of AGI, where tech experts are also divided on its goal, the road to AGI is still hazy.

But most recently, Cognition AI just introduced Devin, an autonomous coding agent that is able to complete tasks. Not just that, it is able to autocomplete tasks and write entire apps on its own in minutes. Developers on X have been concerned about losing their jobs and aiming to brush up on their skills to compete with this massive breakthrough.

Meta is also eyeing AGI with the release of Llama 3. Meta chief Mark Zuckerberg has thrown his hat into the ring in the pursuit of AGI. In his recent Instagram post, he announced, “Our long-term vision is to build general intelligence, open-source it responsibly, and make it widely available so everyone can benefit.”

“We had the opportunity to name our AGI, and we called it Devin,” said a user on X. Maybe AGI would wake up in 2029 (maybe sooner) and say “Hey, I’m Devin.” On the other hand, Yann LeCun on a recent podcast with Fridman said, “You turn on a machine and then we have AGI, that’s not going to happen. It is going to be gradual progress.”

The post AGI Coming Soon appeared first on Analytics India Magazine.

OpenAI Releases Transformer Debugger, an Open Source Tool for Analysing AI Models

OpenAI recently released the Transformer Debugger, a tool that provides insights into the workings of transformer models. This tool marks a step towards greater transparency in AI operations.

This new development comes against the backdrop of its recent criticisms for not open-sourcing its research, alongside Elon Musk announcing his decision to open-source Grok. However, OpenAI has a handful of open-sourced models, including GPT-2, Whisper, CLIP, Jukebox and Point E.

The Transformer Debugger allows for the analysis of transformers’ internal structure. It combines automated interpretability functions and sparse autoencoder technology. This combination facilitates rapid exploration of models, enabling users to understand various aspects of the model’s internal ‘circuitry’ without needing to write code.

The tool is designed to handle neural network components such as neurons and attention heads, offering a practical approach to intervene in the model’s forward pass. For example, users can remove a specific neuron to observe the impact on the model’s output. This feature provides a straightforward method to manually explore and understand the ‘circuitry’ within neural networks, where ‘circuits’ refer to the specific functional components and their interconnections.

Jan Leike, machine learning & alignment researcher at Open AI, said that this research tool is still in its early stages, but, “We are releasing it to let others play with and build on it!” It aims to help researchers uncover why small AI language models behave in certain ways, offering a detailed view of the AI’s decision-making process.

The tool builds on foundational research, including studies on how language models can explain neurons and mono semantic features within language models. However, OpenAI notes that this release does not accompany new findings but rather provides a platform for ongoing exploration and understanding of AI models.

The post OpenAI Releases Transformer Debugger, an Open Source Tool for Analysing AI Models appeared first on Analytics India Magazine.

Infosys Funds Meta’s Llama 2 Project with 22 Indian Languages

Infosys Funds Llama 2 Project with 22 Indian Languages

Yann LeCun, the chief of Meta AI was recently on a podcast with Lex Fridman talking about Meta, AI, and AGI. He revealed that he recently met with an Infosys founder who is funding a project based on Llama 2.

“He’s funding a project to fine-tune Llama 2, the open source model produced by Meta so that it speaks all 22 official languages in India. It’s very important for people in India,” said LeCun. He did not reveal which co-founder he met.

He also spoke about Moustapha Cisse, who used to be a scientist at Meta FAIR who is also using Llama 2 to build language models in Africa, and catering it for medical use.

LeCun is a big proponent of open source AI models and recently also posted on X about the training infrastructure of Llama 3, the upcoming AI model of Meta which is expected to be open source as well.

Infosys was one of the first companies that had invested in OpenAI back in 2015. It donated around USD 1 billion to OpenAI, which was a non-profit back then. Infosys, along with Elon Musk, AWS, YC Research, among others, joined hands to make the USD 1 billion donation to OpenAI.

Apart from Infosys, a lot of Indian companies and startups have been leveraging Llama 2 and making Indic models such as Tamil Llama, Kannada Llama, Telugu Llama, and many more. IIT Bombay’s in work BharatGPT project is also focusing on catering to the 22 Indic languages and is pushing towards being open source.

The post Infosys Funds Meta’s Llama 2 Project with 22 Indian Languages appeared first on Analytics India Magazine.

How Canva is Painting Pixels with GenAI

In October last year, Canva introduced Magic Studio, a suite of generative AI-powered design tools that makes content creation accessible to many users and automates labour-intensive features, regardless of their design experience.

“Magic Studio works on internally-developed AI and ML algorithms that leverage a combination of foundational AI models from our team, including Kaleido, and a variety of partners like OpenAI, Google, AWS, and Runway,” Danny Wu, head of AI products at Canva, told AIM.

The platform introduces several new features, including Magic Switch, enabling users to instantly convert designs into different formats without manual adjustments. Most importantly, it can automatically translate designs into over 100 languages, including Hindi and Bengali. Magic Media also got an upgrade from text-to-image to text-to-video capabilities powered by Runway AI.

Additionally, Magic Studio presents new photo editing tools like Magic Grab for automatic subject selection and separation and Magic Expand for expanding images beyond their frames, comparable to Adobe’s Generative Expand tool for Photoshop.

“Generative AI is core to our product vision and strategy, and we expect that it will fuel every step of the creation and editing process, making it easy for anyone to bring the idea in their head to life,” said Wu, highlighting the democratisation of AI.

Big Bets on Generative AI

“We’ve been early adopters of technology like web-based tools (instead of desktop apps) and leveraging the cloud, the shift to mobile apps, and, of course, AI,” Wu added. The company’s growth trajectory has been marked by a shift from a consumer-centric focus to accommodating the evolving needs of teams and enterprises.

With over 170 million users, the company reached unicorn status in 2018, with a $40 million investment valued at $1 billion. In May 2022, Walt Disney CEO Bob Iger became an angel investor in the Blackbird Ventures and Sequoia Capital-backed company.

In 2022, it took a leap of faith and entered the AI race, offering stiff resistance to competitors like Microsoft Designer and Adobe. Although Wu could not share specific investment details, he added that the ​​design studio’s investment in generative AI aligns with its long-term business goals and revenue projections.

Over the past six years, it has steadily built its AI capabilities and made major acquisitions, such as Kaleido, in 2021. The Vienna-based company came to fame with remove.bg in 2019. This jump-started the integration of AI into Canva’s toolkit with the subsequent launch of Magic Write and Text to Image in 2022, followed by Magic Studio in 2023.

Beyond technology, Wu noted that the team is also “investing heavily in content, our creator community, and developers”, with initiatives like a $200 million fund for creators and content over three years, announced in October 2023. This fund provides financial support and incorporates royalties for creators, compensating them for using their content in AI training.

“We’re also giving our creators a choice on how their data is used — we’re not training GenAI models on their private design content without their consent,” commented Wu.

Additionally, the company has introduced a $50 million Developer Innovation Fund to support and incentivise developers financially.

India is a major market for the company. Its recent survey indicates a growing interest among the nation’s CIOs, with 93% planning increased AI investment.

Considering the survey’s revelation that 55% of CIOs lack sufficient IT staff for training on new apps, Wu believes that Magic Studio’s user-friendliness for non-technical users can drive growth.

Replace Human Artists?

Technology has simplified art creation, allowing anyone to produce impressive images effortlessly. The history of art has seen a transition from traditional tools like paint brushes to digital mediums. AI systems, trained on a vast collection of human-made images, now serve as the foundation for future artistic endeavours.

However, concerns about AI replacing human creativity are also rising simultaneously, leading to a tussle between AI and artists. The core discussion has been whether AI-generated art, lacking human control, truly qualifies as art and, most importantly, will AI replace traditional artists.

“We believe that human creativity is at the core of design. AI will take it to new heights, not replace it. Our AI tools and features are designed to supercharge the creative process,” Wu pointed out.

The company believes that AI, including generative AI, will augment human creativity rather than replace it.

The emphasis is on how AI can enhance artists’ capabilities, enabling faster iteration, idea generation, and collaboration across disciplines. Canva believes human input remains crucial, as human users drive the initial vision and creative choices. The Creator Compensation Program is in line with this vision.

Looking ahead, Wu pointed out that the team wants to add more features for seamless visual communication using AI. The ongoing expansion of Magic Studio is geared towards addressing the requirements of professional users, enabling them to create on-brand visual content efficiently.

The post How Canva is Painting Pixels with GenAI appeared first on Analytics India Magazine.

Meta is Training Llama 3 on 24k NVIDIA H100 Clusters

Meta is Training Llama 3 on 24k NVIDIA H100 Clusters

Meta has unveiled crucial details about its cutting-edge hardware infrastructure, specifically tailored for AI training and as Yann LeCun pointed out for training Llama 3. The company revealed insights into its 24,576-GPU data centre-scale clusters, integral to supporting current and forthcoming AI models, including Llama 3, the successor to Llama 2.

Representing a significant investment in AI hardware, Meta’s clusters underscore the pivotal role of infrastructure in shaping the future of AI. These clusters are designed to power Meta’s long-term vision of creating AGI in an open and responsible manner, aiming for widespread accessibility.

The computing infrastructure for Llama-3 training. https://t.co/xpUfvCHjW0

— Yann LeCun (@ylecun) March 12, 2024

In the latest development, Meta has deployed two variants of its 24,576-GPU clusters, each equipped with distinct network fabric solutions. One cluster utilises a remote direct memory access (RDMA) over converged Ethernet (RoCE) network fabric, while the other features an NVIDIA Quantum2 InfiniBand fabric. Both solutions boast 400 Gbps endpoints, enabling seamless interconnectivity for large-scale training tasks.

Notably, the company’s AI Research SuperCluster (RSC), introduced in 2022, featuring 16,000 NVIDIA A100 GPUs, has been pivotal in advancing open and responsible AI research, facilitating the development of advanced AI models such as Llama and Llama 2.

Through meticulous network, software, and model architecture co-design, Meta has successfully harnessed the capabilities of both RoCE and InfiniBand clusters, mitigating network bottlenecks in large-scale AI workloads. This includes ongoing training sessions of Llama 3 on Meta’s RoCE cluster, demonstrating the effectiveness of the infrastructure in supporting advanced AI training tasks.

Looking ahead to the end of 2024, Meta’s goal is to further expand its infrastructure footprint, encompassing 350,000 NVIDIA H100s. This expansion is part of a comprehensive portfolio initiative aimed at achieving computational capabilities equivalent to nearly 600,000 H100s.

The post Meta is Training Llama 3 on 24k NVIDIA H100 Clusters appeared first on Analytics India Magazine.

AI fraud detection software maker Inscribe.ai lays off 40% of staff

AI fraud detection software maker Inscribe.ai lays off 40% of staff Sarah Perez @sarahintampa / 5 days

Yet another AI-powered fraud detection software provider is laying off staff. Inscribe, whose platform works to detect fraud in areas like business underwriting, tenant screening and onboarding, has cut just under 40% of its staff, which equates to dozens of employees. The news follows that of another small round of layoffs at AI-powered plagiarism detector Turnitin, whose CEO had last year touted how AI would enable the company to reduce its headcount.

According to sources, Inscribe’s board recommended the cuts as the current market caused the startup to miss its revenue goals for over a year.

San Francisco-based Inscribe.ai confirmed the headcount reduction to TechCrunch, noting that the AI advances in the financial services industry necessitated a pivot to a new product and direction for the company.

“2023 was a year of change for our customers and Inscribe,” explained Inscribe CEO and co-founder, Ronan Burke. “Many of our customers in the fintech industry had to contend with higher interest rates and an unpredictable future for consumers and businesses. Additionally, the advances in AI in 2023 present one of the largest opportunities for the financial services ecosystem — enabling improved customer experiences, more efficient processes and fairer decisions,” he continued.

“In Q4 of last year, we set out on a new product strategy to align with these two industry shifts, and we have a large product launch planned for later this year related to this, which we are very excited about. As part of the change in strategy, in January of this year, we made the difficult decision to reduce the size of the team by just under 40%, mostly in go-to-market and operational roles,” Burke said.

The company was already a relatively small operation, with 60 (or more) employees, according to LinkedIn and PitchBook, a mix of engineering, product design, AI expertise, marketing, sales and more.

In January 2023, Inscribe raised $25 million in Series B funding led by Threshold Ventures with participation from Crosslink Capital, Foundry, Uncork Capital, Box co-founder Dillon Smith and Intercom co-founder Des Traynor. The round brought Inscribe’s total raise to date to $38 million. At the time, the company forecast it would double its then 50-person workforce over the coming 12 to 18 months.

Sarah Perez is reachable at sarahp@techcrunch.com or @sarahperez.01 / 415.234.3994 on Signal.

Inscribe bags $25M to fight financial fraud with AI