TOPIC: LINKEDIN
A round-up of online portals for those seeking work
5th August 2025For me, much of 2025 was spent finding a new freelance work engagement. Recently, that search successfully concluded, but not before I got flashbacks of how hard things were when seeking work after completing university education and deciding to hybridise my search to include permanent employment too. Now that I am fulfilling a new contract with a new client, I am compiling a listing of places on the web to a search for work, at least for future reference if nothing else.
Founded in 2011 by former executives from Gumtree, eBay and Zoopla, this UK-based job search engine aggregates listings from thousands of sites across 16+ countries with headquarters in London and approximately 100 employees worldwide. The platform offers over one million job advertisements in the UK alone and an estimated 350 million globally, attracting more than 10 million monthly visits. Jobseekers can use the service without cost, benefiting from search functionality, email alerts, salary insights and tools such as ValueMyCV and the AI-powered interview preparation tool Prepper. The company operates on a Cost-Per-Click or Cost-Per-Applicant model for employers seeking visibility, while also providing data and analytics APIs for programmatic advertising and labour market insights. Notably, the platform powers the UK government Number 10 Dashboard, with its dataset frequently utilised by the ONS for real-time vacancy tracking.
Founded in 2000 by Lee Biggins, this independent job board has grown to become one of the leading platforms in the UK job market. Based in Fleet, Hampshire, it maintains a substantial database of approximately 21.4 million CV's, with around 360,000 new or updated profiles added monthly. The platform attracts significant traffic with about 10.1 million monthly visits from 4.3 million unique users, facilitating roughly 3 million job applications each month across approximately 137,000 live vacancies. Jobseekers can access all services free of charge, including job searching, CV uploads, job alerts and application tracking, though the CV building tools are relatively basic compared to specialist alternatives. The platform boasts high customer satisfaction, with 96 percent of clients rating their service as good or excellent, and offers additional value through its network of over 800 partner job sites and ATS integration capabilities.
Formerly known as TryRemotely, Empllo functions as a comprehensive job board specialising in remote technology and startup positions across various disciplines including engineering, product, sales, marketing, design and finance. The platform currently hosts over 30,000 active listings from approximately 24,000 hiring companies worldwide, with specific regional coverage including around 375 positions in the UK and 36 in Ireland. Among its notable features is the AI-powered Job Copilot tool, which can automatically apply to roles based on user preferences. While Empllo offers extensive listings and advanced filtering options by company, funding and skills, it does have limitations including inconsistent salary information and variable job quality. The service is free to browse, with account creation unlocking personalised features. It is particularly suitable for technology professionals seeking distributed work arrangements with startups, though users are advised to verify role details independently and potentially supplement their search with other platforms offering employer reviews for more thorough vetting.
This is a comprehensive job-hunt management tool that replaces traditional spreadsheets with an intuitive Kanban board interface, allowing users to organise their applications effectively. The platform features a Chrome extension that integrates with major job boards like LinkedIn and Indeed, enabling one-click saving of job listings. Users can track applications through various stages, store relevant documents and contact information, and access detailed statistics about their job search progress. The service offers artificial intelligence capabilities powered by GPT-4 to generate application responses, personalise cover letters and craft LinkedIn profiles. With over 25,000 active users who have tracked more than 280,000 job applications collectively, the tool provides both free and premium tiers. The basic free version includes unlimited tracking of applications, while the Pro subscription adds features such as custom columns, unlimited tags and expanded AI capabilities. This solution particularly benefits active jobseekers managing numerous applications across different platforms who desire structured organisation and data-driven insights into their job search.
This organisation provides a specialised platform matching candidates with companies based on flexible working arrangements, including remote options, location independence and customisable hours. Their interface features a notable "Work From Anywhere" filter highlighting roles with genuine location flexibility, alongside transparency scores for companies that reflect their openness regarding working arrangements. The platform allows users to browse companies offering specific perks like part-time arrangements, sabbatical leave, or compressed hours, with rankings based on flexibility and workplace culture. While free to use with job-saving capabilities and quick matching processes, it appears relatively new with a modest-sized team, limited independent reviews and a smaller volume of job listings compared to more established competitors. The platform's distinctive approach prioritises work-life balance through values-driven matching and company-oriented filters, particularly useful for those seeking roles aligned with modern flexible working preferences.
Founded in 2007 and based in Puerto Rico, FlexJobs operates as a subscription-based platform specialising in remote, hybrid, freelance and part-time employment opportunities. The service manually verifies all job listings to eliminate fraudulent postings, with staff dedicating over 200 hours daily to screening processes. Users gain access to positions across 105+ categories from entry-level to executive roles, alongside career development resources including webinars, resume reviews and skills assessments. Pricing options range from weekly trials to annual subscriptions with a 30-day money-back guarantee. While many users praise the platform for its legitimacy and comprehensive filtering tools, earning high ratings on review sites like Trustpilot, some individuals question whether the subscription fee provides sufficient value compared to free alternatives. Potential limitations include delayed posting of opportunities and varying representation across different industries.
Founded in November 2004 and now operating in over 60 countries with 28 languages, this leading global job search platform serves approximately 390 million visitors monthly worldwide. In the UK alone, it attracts about 34 million monthly visits, with users spending nearly 7 minutes per session and viewing over 8.5 pages on average. The platform maintains more than 610 million jobseeker profiles globally while offering free services for candidates including job searching, application tools, CV uploads, company reviews and salary information. For employers, the business model includes pay-per-click and pay-per-applicant sponsored listings, alongside tools such as Hiring Insights providing salary data and application trends. Since October 2024, visibility for non-sponsored listings has decreased, requiring employers to invest in sponsorship for optimal visibility. Despite this competitive environment requiring strategic budget allocation, the platform remains highly popular due to its comprehensive features and extensive reach.
A meta-directory founded in 2022 by Rodrigo Rocco, this platform aggregates and organises links to over 400 specialised and niche job sites across various industries and regions. Unlike traditional job boards, it does not host listings directly but serves as a discovery tool that redirects users to external platforms where actual applications take place. The service refreshes links approximately every 45 minutes and offers a weekly newsletter. While providing free access and efficient discovery of relevant boards by category or sector, potential users should note that the platform lacks direct job listings, built-in application tracking, or alert systems. It is particularly valuable for professionals exploring highly specialised fields, those wishing to expand beyond mainstream job boards and recruiters seeking to increase their visibility, though beginners might find navigating numerous destination boards somewhat overwhelming.
Founded in Milan by Vito Lomele in 2006 (initially as Jobespresso), this global job aggregator operates in 58 countries and 21 languages. The platform collects between 28 and 35 million job listings monthly from various online sources, attracting approximately 55 million visits and serving over 100 million registered users. The service functions by gathering vacancies from career pages, agencies and job boards, then directing users to original postings when they search. For employers, it offers programmatic recruitment solutions using artificial intelligence and taxonomy to match roles with candidates dynamically, including pay-per-applicant models. While the platform benefits from its extensive global reach and substantial job inventory, its approach of redirecting to third-party sites means the quality and freshness of listings can vary considerably.
Founded in 1993 as Fax-Me Ltd and rebranded in 1995, this pioneering UK job board launched the world's first jobs-by-email service in May 1994. Originally dominating the IT recruitment sector with up to 80% market share in the early 2000s, the platform published approximately 200,000 jobs and processed over 1 million applications monthly by 2010. Currently headquartered in Colchester, Essex, the service maintains a global presence across Europe, North America and Australia, delivering over 1.2 million job-subscription emails daily. The platform employs a proprietary smart matching engine called Alchemy and features manual verification to ensure job quality. While free for jobseekers who can upload CVs and receive tailored job alerts, employers can post vacancies and run recruitment campaigns across various sectors. Although respected for its legacy and niche focus, particularly in technical recruitment, its scale and visibility are more modest compared to larger contemporary platforms.
Founded in 2020 with headquarters in London, Lifelancer operates as an AI-powered talent hiring platform specialising in life sciences, pharmaceutical, biotech, healthcare IT and digital health sectors. The company connects organisations with freelance, remote and international professionals through services including candidate matching and global onboarding assistance. Despite being relatively small, Lifelancer provides distinct features for both hiring organisations and jobseekers. Employers can post positions tailored to specific healthcare and technology roles, utilising AI-based candidate sourcing, while professionals can create profiles to be matched with relevant opportunities. The platform handles compliance and payroll across multiple countries, making it particularly valuable for international teams, though as a young company, it may not yet offer the extensive talent pool of more established competitors in the industry.
The professional networking was core to my search for work and had its uses while doing so. Writing posts and articles did a lot to raise my profile along with reaching out to others, definitely an asset when assessing the state of a freelancing market. The usefulness of the green "Open to Work" banner is debatable because of my freelancing pitch in a slow market. Nevertheless, there was one headhunting approach that might have resulted in something if another offer had not gazumped it. Also, this is not a place to hang around over a weekend with job search moaning filling your feed, though making your interests known can change that. Now that I have paid work, the platform has become a way of keeping up to date in my line of business.
Established in 1994 as The Monster Board, Monster.com became one of the first online job portals, gaining prominence through memorable Super Bowl advertisements. As of June 2025, the platform attracts approximately 4.3 million monthly visits, primarily from the United States (76%), with smaller audiences in India (6%) and the UK (1.7%). The service offers free resources for jobseekers, including resume uploads and career guidance, while employers pay for job postings and additional premium features.
Established in 1999 and headquartered in Richmond, Surrey, PharmiWeb has evolved into Europe's leading pharmaceutical and life sciences platform. The company separated its dedicated job board as PharmiWeb.jobs in 2019, while maintaining industry news and insights on the original portal. With approximately 600,000 registered jobseekers globally and around 200,000 monthly site visits generating 40,000 applications, the platform hosts between 1,500 and 5,000 active vacancies at any time. Jobseekers can access the service completely free, uploading CVs and setting alerts tailored to specific fields, disciplines or locations. Additional recruiter services include CV database access, email marketing campaigns, employer branding and applicant management tools. The platform particularly excels for specialised pharmaceutical, biotech, clinical research and regulatory affairs roles, though its focused nature means it carries fewer listings than mainstream employment boards and commands higher posting costs.
If 2025 was a flashback to the travails of seeking work after completing university education, meeting this name again was another part of that. Founded in May 1960 by Sir Alec Reed, the firm began as a traditional recruitment agency in Hounslow, West London, before launching the first UK recruitment website in 1995. Today, the platform attracts approximately 3.7 million monthly visitors, primarily UK-based users aged 25-34, generating around 80,000 job applications daily. The service offers jobseekers free access to search and apply for roles, job alerts, CV storage, application tracking, career advice articles, a tax calculator, salary tools and online courses. For employers, the privately owned company provides job advertising, access to a database of 18-22 million candidate CVs and specialist recruitment across about 20 industry sectors.
Founded by digital nomad Pieter Levels in 2015, this prominent job board specialises exclusively in 100% remote positions across diverse sectors including tech, marketing, writing, design and customer support. The platform offers free browsing and application for jobseekers, while employers pay fees. Notable features include mandatory salary transparency, global job coverage with regional filtering options and a clean, minimalist interface that works well on mobile devices. Despite hosting over 100,000 remote jobs from reputable companies like Amazon and Microsoft, the platform has limitations including basic filtering capabilities and highly competitive application processes, particularly for tech roles. The simple user experience redirects applications directly to employer pages rather than using an internal system. For professionals seeking remote work worldwide, this board serves as a valuable resource but works best when used alongside other specialised platforms to maximise opportunities.
Founded in 2015 and based in Boulder, Colorado, this platform exclusively focuses on remote work opportunities across diverse industries such as marketing, finance, healthcare, customer support and design. Attracting over 1.5 million monthly visitors, it provides jobseekers with free access to various employment categories including full-time, part-time, freelance and hybrid positions. Beyond job listings, the platform offers a comprehensive resource centre featuring articles, expert insights and best practices from over 108 remote-first companies. Job alerts and weekly newsletters keep users informed about relevant opportunities. While the platform provides strong resources and maintains positive trust ratings of approximately 4.2/5 on Trustpilot, its filtering capabilities are relatively basic compared to competitors. Users might need to conduct additional research as company reviews are not included with job postings. Despite these limitations, the platform serves as a valuable resource for individuals seeking remote work guidance and opportunities.
For jobseekers in the technology and digital sectors, Remotive serves as a specialised remote job board offering approximately 2,000 active positions on its free public platform. Founded around 2014-2015, this service operates with a remote-first approach and focuses on verifying job listings for legitimacy. The platform provides a premium tier called "Remotive Accelerator" which grants users access to over 50,000 additional curated jobs, advanced filtering options based on skills and salary requirements and membership to a private Slack community. While the interface receives praise for its clean design and intuitive navigation, user feedback regarding the paid tier remains mixed, with some individuals noting limitations such as inactive community features and an abundance of US-based or senior-level positions. The platform is particularly valuable for professionals in software development, product management, marketing and customer service who are seeking global remote opportunities.
Originally launched in Canada in 2011 as neuvoo, this global job search engine is now headquartered in Montreal, Quebec, providing access to over 30 million jobs across more than 75 countries. The platform attracts between 12 and 16 million monthly visits worldwide, with approximately 6 percent originating from the UK. Jobseekers can utilise the service without charge, accessing features like salary converters and tax calculators in certain regions to enhance transparency about potential earnings. Employers have the option to post jobs for free in some areas, with additional pay per click sponsored listings available to increase visibility. Despite its extensive coverage and useful tools, user feedback remains mixed, with numerous complaints on review sites regarding outdated listings, unwanted emails and difficulties managing or deleting accounts.
Founded in 1999, Totaljobs is a major UK job board currently owned by StepStone Group UK Ltd, a subsidiary of Axel Springer Digital Classifieds. The platform attracts approximately 20 million monthly visits and generates 4-5 million job applications each month, with over 300,000 daily visitors browsing through typically 280,000+ live job listings. As the flagship of a broader network including specialised boards such as Jobsite, CareerStructure and City Jobs, Totaljobs provides jobseekers with search functionality across various sectors, job alerts and career advice resources. For employers and recruiters, the platform offers pay-per-post job advertising, subscription options for CV database access and various employer tools.
Founded in 2011, this is one of the largest purely remote job boards globally, attracting approximately 6 million monthly visitors and featuring over 36,000 remote positions across various categories including programming, marketing, customer support and design. Based in Vancouver, the platform operates with a small remote-first team who vet listings to reduce spam and scams. Employers pay for each standard listing, while jobseekers access the service without charge. The interface is straightforward and categorised by functional area, earning trust from major companies like Google, Amazon and GitHub. However, the platform has limitations including basic filtering capabilities, a predominance of senior-level positions particularly in technology roles and occasional complaints about outdated or misleading posts. The service is most suitable for experienced professionals seeking genuine remote opportunities rather than those early in their careers. Some users report region-restricted application access and positions that offer lower compensation than expected for the required experience level.
Founded in 2014, this job board provides remote work opportunities for digital nomads and professionals across various industries. The platform offers over 30,000 fully remote positions spanning sectors such as technology, marketing, writing, finance and education. Users can browse listings freely, but a Premium subscription grants access to additional jobs, enhanced filters and email alerts. The interface is user-friendly with fast-loading pages and straightforward filtering options. The service primarily features global employment opportunities suitable for location-independent workers. However, several limitations exist: many positions require senior-level experience, particularly in technical fields; the free tier displays only a subset of available listings; filtering capabilities are relatively basic; and job descriptions sometimes lack detail. The platform has received mixed reviews, earning approximately 3.4 out of 5 on Trustpilot, with users noting the prevalence of senior technical roles and questioning the value of the premium subscription. It is most beneficial for experienced professionals comfortable with remote work arrangements, while those seeking entry-level positions might find fewer suitable opportunities.
From boardroom to code: More options for AI and Data Science education
27th July 2025The artificial intelligence revolution has created an unprecedented demand for education that spans from executive strategy to technical implementation. Modern professionals face the challenge of navigating a landscape where understanding AI's business implications proves as crucial as mastering its technical foundations. This comprehensive examination explores five distinguished programmes that collectively address this spectrum, offering pathways for business professionals, aspiring data scientists and technical specialists seeking advanced expertise.
Strategic Business Implementation Through Practical AI Tools
LinkedIn Learning's Applying Generative AI as a Business Professional programme represents the entry point for professionals seeking immediate workplace impact. This focused five-hour curriculum across six courses addresses the practical reality that most business professionals need functional AI literacy rather than technical mastery. The programme emphasises hands-on application of contemporary tools including ChatGPT, Claude and Microsoft Copilot, recognising that these platforms have become integral to modern professional workflows.
The curriculum's strength lies in its emphasis on prompt engineering techniques that yield immediate productivity gains. Participants learn to craft effective queries that consistently produce useful outputs, a skill that has rapidly evolved from novelty to necessity across industries. The programme extends beyond basic tool usage to include strategies for creating custom GPTs without programming knowledge, enabling professionals to develop solutions that address specific organisational challenges.
Communication enhancement represents another critical component, as the programme teaches participants to leverage AI for improving written correspondence, presentations and strategic communications. This practical focus acknowledges that AI's greatest business value often emerges through augmenting existing capabilities rather than replacing human expertise. The inclusion of critical thinking frameworks for AI-assisted decision-making ensures that participants develop sophisticated approaches to integrating artificial intelligence into complex business processes.
Academic Rigour Meets Strategic AI Governance
The University of Pennsylvania's AI for Business Specialisation on Coursera elevates business AI education to an academic level whilst maintaining practical relevance. This four-course programme, completed over approximately four weeks, addresses the strategic implementation challenges that organisations face when deploying AI technologies at scale. The curriculum's foundation in Big Data fundamentals provides essential context for understanding the data requirements that underpin successful AI initiatives.
The programme's exploration of machine learning applications in marketing and finance demonstrates how AI transforms traditional business functions. Participants examine customer journey optimisation techniques, fraud prevention methodologies and personalisation technologies that have become competitive necessities rather than optional enhancements. These applications receive thorough treatment that balances technical understanding with strategic implications, enabling participants to make informed decisions about AI investments and implementations.
Particularly valuable is the programme's emphasis on AI-driven people management practices, addressing how artificial intelligence reshapes human resources, talent development and organisational dynamics. This focus acknowledges that successful AI implementation requires more than technological competence; it demands sophisticated understanding of how these tools affect workplace relationships and employee development.
The specialisation's coverage of strategic AI governance frameworks proves especially relevant as organisations grapple with ethical deployment challenges. Participants develop comprehensive approaches to responsible AI implementation that address regulatory compliance, bias mitigation and stakeholder concerns. This academic treatment of AI ethics provides the foundational knowledge necessary for creating sustainable AI programmes that serve both business objectives and societal responsibilities.
Industry-Standard Professional Development
IBM's Data Science Professional Certificate represents a bridge between business understanding and technical proficiency, offering a comprehensive twelve-course programme designed for career transition. This four-month pathway requires no prior experience whilst building industry-ready capabilities that align with contemporary data science roles. The programme's strength lies in its integration of technical skill development with practical application, ensuring graduates possess both theoretical knowledge and hands-on competency.
The curriculum's progression from Python programming fundamentals through advanced machine learning techniques mirrors the learning journey that working data scientists experience. Participants gain proficiency with industry-standard tools including Jupyter notebooks, GitHub and Watson Studio, ensuring familiarity with the collaborative development environments that characterise modern data science practice. This tool proficiency proves essential for workplace integration, as contemporary data science roles require seamless collaboration across technical teams.
The programme's inclusion of generative AI applications reflects IBM's recognition that artificial intelligence has become integral to data science practice rather than a separate discipline. Participants learn to leverage AI tools for data analysis, visualisation and insight generation, developing capabilities that enhance productivity whilst maintaining analytical rigour. This integration prepares trainees for data science roles that increasingly incorporate AI-assisted workflows.
Real-world project development represents a crucial component, as participants build comprehensive portfolios that demonstrate practical proficiency to potential employers. These projects address authentic business challenges using genuine datasets, ensuring that participants can articulate their capabilities through concrete examples.
Advanced Technical Mastery Through Academic Excellence
Andrew Ng's Machine Learning Specialisation on Coursera establishes the technical foundation for advanced AI practice. This three-course programme, completed over approximately two months, provides comprehensive coverage of core machine learning concepts whilst emphasising practical implementation skills. Andrew Ng's reputation as an AI pioneer lends exceptional credibility to this curriculum, ensuring that participants receive instruction that reflects both academic rigour and industry best practices.
The specialisation's treatment of supervised learning encompasses linear and logistic regression, neural networks and decision trees, providing thorough grounding in the algorithms that underpin contemporary machine learning applications. Participants develop practical proficiency with Python, NumPy and scikit-learn, gaining hands-on experience with the tools that professional machine learning practitioners use daily. This implementation focus ensures that theoretical understanding translates into practical capability.
Unsupervised learning includes clustering algorithms, anomaly detection techniques and certain approaches in recommender systems, all of which contribute to powering modern digital experiences. The programme's exploration of reinforcement learning provides exposure to the techniques driving advances in autonomous systems and game-playing AI. This breadth ensures that participants understand the full spectrum of machine learning approaches, rather than developing narrow expertise in specific techniques.
Cutting-Edge Deep Learning Applications
Again available through Coursera, Andrew Ng's Deep Learning Specialisation extends technical education into the neural network architectures that drives contemporary AI. This five-course programme, spanning approximately three months, addresses the advanced techniques that enable computer vision, natural language processing and complex pattern recognition applications. The intermediate-level curriculum assumes foundational machine learning knowledge whilst building expertise in cutting-edge methodologies.
Convolutional neural network coverage provides comprehensive understanding of computer vision applications, from image classification through object detection and facial recognition. Participants develop practical skills with CNN architectures that power visual AI applications across industries. The programme's treatment of recurrent neural networks and LSTMs addresses sequence processing challenges in speech recognition, machine translation and time series analysis.
The specialisation's exploration of transformer architectures proves particularly relevant given their central role in large language models and natural language processing breakthroughs. Participants gain understanding of attention mechanisms, transfer learning techniques and the architectural innovations that enable modern AI capabilities. This coverage ensures they understand the technical foundations underlying contemporary AI advances.
Real-world application development represents a crucial component, as participants work on speech recognition systems, machine translation applications, image recognition tools and chatbot implementations. These projects utilise TensorFlow, a dominant framework for deep learning development, ensuring that graduates possess practical experience with production-ready tools.
Strategic Integration and Future Pathways
These five programmes collectively address the comprehensive skill requirements of the modern AI landscape, from strategic business implementation through advanced technical development. The progression from practical tool usage through academic business strategy to technical mastery reflects the reality that successful AI adoption requires capabilities across multiple domains. Organisations benefit most when business leaders understand AI's strategic implications, whilst technical teams possess sophisticated implementation capabilities.
The integration of business strategy with technical education acknowledges that artificial intelligence's transformative potential emerges through thoughtful application rather than technological sophistication alone. These programmes prepare professionals to contribute meaningfully to AI initiatives regardless of their specific role or technical background, ensuring that organisations can build comprehensive AI capabilities that serve both immediate needs and long-term strategic objectives.
From mathematical insights to practical applications: Two perspectives on AI
19th April 2025As AI continues to transform our technological landscape, two recent books offer distinct yet complementary perspectives on understanding and working with these powerful tools. Stephen Wolfram's technical deep dive and Ethan Mollick's practical guide approach the subject from different angles, but both provide valuable insights for navigating our AI-integrated future.
What is ChatGPT Doing?: Wolfram's Technical Lens
Stephen Wolfram's exploration of large language models is characteristically thorough and mathematically oriented. While dense in parts, his analysis reveals fascinating insights about both AI and human cognition.
Perhaps most intriguing is Wolfram's observation that generative AI unexpectedly teaches us about human language production. These systems, in modelling our linguistic patterns with such accuracy, hold up a mirror to our own cognitive processes, perhaps revealing structures and patterns we had not fully appreciated before.
Wolfram does not shy away from highlighting limitations, particularly regarding computational capabilities. As sophisticated as next-word prediction has become through multi-billion parameter neural networks, these systems fundamentally lack true mathematical reasoning. However, his proposal of integrating language models with computational tools like WolframAlpha presents an elegant solution, combining the conversational fluency of AI with precise computational power.
Co-intelligence: Mollick's Practical Framework
Ethan Mollick takes a decidedly more accessible approach in "Co-intelligence," offering accessible strategies for effective human-AI collaboration across various contexts. His framework includes several practical principles:
- Invite AI to the table as a collaborator rather than merely a tool
- Maintain human oversight and decision-making authority
- Communicate with AI systems as if they were people with specific roles
- Assume current AI represents the lowest capability level you will work with going forward
What makes Mollick's work particularly valuable is its contextual applications. Drawing from his background as a business professor, he methodically examines how these principles apply across different collaborative scenarios: from personal assistant to creative partner, coworker, tutor, coach, and beyond. With a technology, that, even now, retains some of the quality of a solution looking for a problem, these grounded suggestions act as a counterpoint to the torrent of hype that that deluges our working lives, especially if you frequent LinkedIn a lot as I am doing at this time while searching for new freelance work.
Complementary Perspectives
Though differing significantly in their technical depth and intended audience, both books contribute meaningfully to our understanding of AI. Wolfram's mathematical rigour provides theoretical grounding, while Mollick's practical frameworks offer immediate actionable insights. For general readers looking to productively integrate AI into their work and life, Mollick's accessible approach serves as an excellent entry point. Those seeking deeper technical understanding will find Wolfram's analysis challenging but rewarding.
As we navigate this rapidly evolving landscape, perspectives from both technical innovators and practical implementers will be essential in helping us maximise the benefits of AI while mitigating potential drawbacks. As ever, the hype outpaces the practical experiences, leaving us to suffer the marketing output while awaiting real experiences to be shared. It is the latter is more tangible and will allow us to make use of game-changing technical advances.
Finding human balance in an age of AI code generation
12th March 2025Recently, I was asked about how I felt about AI. Given that the other person was not an enthusiast, I picked on something that happened to me, not so long ago. It involved both Perplexity and Google Gemini when I was trying to debug something: both produced too much code. The experience almost inspired a LinkedIn post, only for some of the thinking to go online here for now. A spot of brainstorming using an LLM sounds like a useful exercise.
Going back to the original question, it happened during a meeting about potential freelance work. Thus, I tapped into experiences with code generators over several decades. The first one involved a metadata-driven tool that I developed; users reported that there was too much imperfect code to debug with the added complexity that dealing with clinical study data brings. That challenge resurfaced with another bespoke tool that someone else developed, and I opted to make things simpler: produce some boilerplate code and let users take things from there. Later, someone else again decided to have another go, seemingly with more success.
It is even more challenging when you are insufficiently familiar with the code that is being produced. That happened to me with shell scripting code from Google Gemini that was peppered with some Awk code. There was no alternative but to learn a bit more about the language from Tutorials Point and seek out an online book elsewhere. That did get me up to speed, and I will return to these when I am in need again.
Then, there was the time when I was trying to get a Julia script to deal with Google Drive needing permissions to be set. This started Google Gemini into adding more and more error checking code with try catch blocks. Since I did not have the issue at that point, I opted to halt and wait for its recurrence. When it did, I opted for a simpler approach, especially with the gdrive CLI tool starting up a web server for completing the process of reactivation. While there are times when shell scripting is better than Julia for these things, I added extra robustness and user-friendliness anyway.
During that second task, I was using VS Code with the GitHub Copilot plugin. There is a need to be careful, yet that can save time when it adds suggestions for you to include or reject. The latter may apply when it adds conditional logic that needs more checking, while simple code outputting useful text to the console can be approved. While that certainly is how I approach things for now, it brings up an increasingly relevant question for me.
How do we deal with all this code production? In an environment with myriads of unit tests and a great deal of automation, there may be more capacity for handling the output than mere human inspection and review, which can overwhelm the limitations of a human context window. A quick search revealed that there are automated tools for just this purpose, possibly with their own learning curves; otherwise, manual working could be a better option in some cases.
After all, we need to do our own thinking too. That was brought home to me during the Julia script editing. To come up with a solution, I had to step away from LLM output and think creatively to come up with something simpler. There was a tension between the two needs during the exercise, which highlighted how important it is to learn not to be distracted by all the new technology. Being an introvert in the first place, I need that solo space, only to have to step away from technology to get that when it was a refuge in the first place.
For anyone with a programming hobby, they have to limit all this input to avoid being overwhelmed; learning a programming language could involve stripping out AI extensions from a code editor, for instance, LLM output has its place, yet it has to be at a human scale too. That perhaps is the genius of a chat interface, and we now have Agentic AI too. It is as if the technology curve never slackens, at least not until the current boom ends, possibly when things break because they go too far beyond us. All this acceleration is fine until we need to catch up with what is happening.
Broadening data science horizons: Useful Python packages for working with data
14th October 2021My response to changes in the technology stack used in clinical research is to develop some familiarity with programming and scripting platforms that complement and compete with SAS, a system with which I have been programming since 2000. While one of these has been R, Python is another that has taken up my attention, and I now also have Julia in my sights as well. There may be others to assess in the fullness of time.
While I first started to explore the Data Science world in the autumn of 2017, it was in the autumn of 2019 that I began to complete LinkedIn training courses on the subject. Good though they were, I find that I need to actually use a tool to better understand it. At that time, I did get to hear about Python packages like Pandas, NumPy, SciPy, Scikit-learn, Matplotlib, Seaborn and Beautiful Soup though it took until of spring of this year for me to start gaining some hands-on experience with using any of these.
During the summer of 2020, I attended a BCS webinar on the CodeGrades initiative, a programming mentoring scheme inspired by the way classical musicianship is assessed. In fact, one of the main progenitors is a trained classical musician and teacher of classical music who turned to Python programming when starting a family to have a more stable income. The approach is that a student selects a project and works their way through it, with mentoring and periodic assessments carried out in a gentle and discursive manner. Of course, the project has to be engaging for the learning experience to stay the course, and that point came through in the webinar.
That is one lesson that resonates with me with subjects as diverse as web server performance and the ongoing pandemic supplying data, and there are other sources of public data to examine as well before looking through my own personal archive gathered over the decades. Though some subjects are uplifting while others are more foreboding, the key thing is that they sustain interest and offer opportunities for new learning. Without being able to dream up new things to try, my knowledge of R and Python would not be as extensive as it is, and I hope that it will help with learning Julia too.
In the main, my own learning has been a solo effort with consultation of documentation along with web searches that have brought me to the likes of Real Python, Stack Abuse, Data Viz with Python and R and others for longer tutorials as well as threads on Stack Overflow. Usually, the web searching begins when I need a steer on a particular or a way to resolve a particular error or warning message, but books are always worth reading even if that is the slower route. While those from the Dummies series or from O'Reilly have proved must useful so far, I do need to read them more completely than I already have; it is all too tempting to go with the try the "programming and search for solutions as you go" approach instead.
To get going, many choose the Anaconda distribution to get Jupyter notebook functionality, but I prefer a more traditional editor, so Spyder has been my tool of choice for Python programming and there are others like PyCharm as well. Because Spyder itself is written in Python, it can be installed using pip from PyPi like other Python packages. It has other dependencies like Pylint for code management activities, but these get installed behind the scenes.
The packages that I first met in 2019 may be the mainstays for doing data science, but I have discovered others since then. It also seems that there is porosity between the worlds of R and Python, so you get some Python packages aping R packages and R has the Reticulate package for executing Python code. There are Python counterparts to such Tidyverse stables as dplyr and ggplot2 in the form of Siuba and Plotnine, respectively. Though the syntax of these packages are not direct copies of what is executed in R, they are close enough for there to be enough familiarity for added user-friendliness compared to Pandas or Matplotlib. The interoperability does not stop there, for there is SQLAlchemy for connecting to MySQL and other databases (PyMySQL is needed as well) and there also is SASPy for interacting with SAS Viya.
While Python may not have the speed of Julia, there are plenty of packages for working with larger workloads. Of these, Dask, Modin and RAPIDS all have their uses for dealing with data volumes that make Pandas code crawl. As if to prove that there are plenty of libraries for various forms of data analytics, data science, artificial intelligence and machine learning, there also are the likes of Keras, TensorFlow and NetworkX. These are just a selection of what is available, and there is always the possibility of checking out others. It may be tempting to stick with the most popular packages all the time, especially when they do so much, but it never hurts to keep an open mind either.
A new phone
4th February 2012After a few years with a straightforward Nokia 1661 and a PAYG Blackberry 8520, I decided to go and upgrade from the former to an HTC Wildfire S. So far, the new phone has been good to me with only a few drawbacks. Other than working out how to insert a SIM card, the phone has been easy to use with just a few nuances to learn, such as finger pinch zooming and dealing with an onscreen keyboard as opposed to a real one.
The touchscreen and 3G connectivity are major upgrades from my Blackberry, making web browsing much faster, especially on the larger screen. Checking Google Reader and emails on the go is quicker, with the screen responding well most of the time. It does get dirty, so using a screen protector or regularly cleaning with a lens cloth is advisable. As it happens, I'm still adjusting to the onscreen keyboard, which remains the one area where the Blackberry remains superior. Rotating the phone sideways helps by enlarging the keys, reducing typing errors even for my average-sized fingers. Switching between alphabetic, numeric, and punctuation keyboards still takes some getting used to.
Otherwise, the user interface is bright and pleasing to the eye, with the typical presentation of both a clock and current weather on there. Handily, the screen is locked easily too with a press of the button at the top right of the phone. That will put a stop to inadvertent phone calls, emailing, web browsing and other things, so it is to be commended. To unlock the screen, all that's needed is to swipe the lock bar to the bottom. Any alerts are viewed similarly with holding down your finger on the top bar presenting an extension that can be pulled all the way down to see what's there.
The Android Marketplace icon on the home screen lets me easily add apps with automatic updates, though this requires monitoring data usage on your phone plan. The WordPress app works better than on my Blackberry, but UberSocial's retweeting function is worse on Android. It displays all account feeds on one screen and requires swiping for actions like replying or retweeting, which I find awkward. I might try an alternative app. I've downloaded several others, including CrossCountry Trains' app (which is good, despite failing to find Macclesfield-Edale Sunday trains) and LinkedIn (which works well). You can move apps to the microSD card to save internal storage space, though I don't plan to install many.
The Wildfire performs well at its core function: making and receiving calls. It imported contacts from my SIM card, though Bluetooth transfer from an old phone is also possible. Call sound quality is clear and loud. The side rocker button adjusts speaker volume during calls and ringtone volume otherwise. By default, the phone vibrates and rings simultaneously for incoming calls, which I may change later. The same applies to notification sounds for text messages, emails, and tweets.
Battery life is this phone's main weakness. It needs charging every night, unlike my previous phones. The bright, responsive screen likely causes this drain. Many users report similar issues online, with some experiencing even worse battery performance. While there are tips for extending battery life, they involve disabling key features like 3G or data connectivity, which defeats the purpose of having a smartphone. Thus, I'm considering buying a spare battery, as I did for my Pentax DSLR. Some users recommend higher-capacity replacement batteries, though this seems riskier.
All in all, first impressions of the HTC Wildfire are good ones. Over time, I should find out more about the ins and outs of the gadget. After all, it is a mini-computer with its own operating system and other software. Since I continue to learn more and more about PC's every day, the same should be the case here too.
Exploring the mobile web
16th July 2010With a change of job ahead of me, I decided to make my web usage a little more mobile. The result was the purchase of a Blackberry 8520 Curve on a T-Mobile pay-as-you-go tariff to complement my existing phone. The attraction included mobile email access and some web browsing capability. Though GPRS provides slow web browsing, it works adequately whilst highlighting the value of mobile-optimised websites. It's just as well that this website that you're reading has a mobile version.
Hooking the Blackberry up to GMail was no problem once I had paid my dues and the necessary set up was done for me; it was only then that the required option was available through the set-up screens. While RIM's own web browser may be no slouch when it comes to rendering websites, I put Opera Mini in place as well for those times when the default option could be bettered and they exist too. Speaking of RIM applications, there's one for Twitter too, though I added Übertwitter for the sake of greater flexibility (it can handle more than one account at a time, for example). In addition, I have instated applications for WordPress and LinkedIn too, and it was then that I stopped myself spending too much time in Blackberry App World. If I was of the Facebook persuasion, I might be interested in the default offering for that as well, but I have learnt to contain myself.
Of course, there are limitations to the device's capabilities regarding email and web on the move. Long emails still need desktop access (messages can get truncated) and mobile unfriendly websites will take an age to load and explore; a small screen means much more finger work. After all, this is a small device, so the observations aren't astounding; it's just that I encounter the reality of life on a small screen now. Nevertheless, useful sites like those from Google and the Met Office have a mobile variant, though I'd like to see the latter including its rain radar as part of the package.
Speaking of life on a smaller scale, there's the size of the keyboard to consider too. So far, I haven't had much practice with it, but I am unsure as how some craft longer blog entries with the tiny keys. Then, there's the ever-present threat of arm discomfort and RSI that you have to watch. For that reason, I'll stick with use for an hour at a time rather than going mad altogether. Navigating around the screen using the tiny trackpad is something to which I am adjusting, and it works well enough too, so long as you're not looking through long web pages or emails.
To bring this piece to a close, the new gadget has been finding uses and I don't intend to leave it idle after paying over £150 for it. Apart from acting as an expensive calculator, it has already travelled abroad with me with roaming not being a problem; while I may have failed to make it work with hotel broadband, there was EDGE availability to keep things connected. All in all, the device is earning its keep and teaching me a few things about mobile handheld computing with my main website in process of being made more mobile compatible with the front page and the photo gallery gaining versions for handheld devices after the same was done for the outdoors blog earlier this year (might make the design look more like the rest of the site though). Without something on which to do some real testing, that idea may not have become reality as it is. It may be no desktop substitute, but that's never to say that these devices may never get near that situation. After all, there was a time when no one could imagine the same for laptop PC's, and we all know what has happened to them.