Technology Tales

PumasAI is a pharmaceutical technology company based in Dover, Delaware, that develops data analytics tools designed to support drug development and healthcare delivery. Its flagship software, Pumas, has been used in over 26 successful regulatory submissions, and the company claims to have saved its 60-plus clients a combined total of one billion dollars. The firm has recently released version 2.8 of its platform, which includes a new feature called PumasAide, and has been recognised at the Biotechnology Awards for its contributions to the pharmaceutical industry. Alongside its software products, the company offers consulting services aimed at helping clients navigate the regulatory approval process, as well as complimentary access to its modelling tools for those engaged in non-commercial research and education.

The notion that AI will decimate the job market for developers is greatly exaggerated. In fact, AI represents a platform shift that's changing what it looks like to build software and ushering in a period of enormous demand for ambitious, innovative and highly specialised code. This demand is driven by the imagination engine of the human mind, which constantly comes up with better ways of doing things. Each imagined future requires software to become reality, leading to new jobs and new approaches to existing ones. The changing nature of development work means that developers are shifting from writing every line of code by hand to orchestrating AI agents that generate code. New roles are emerging, such as AI orchestrators, prompt engineers and human-AI collaboration architects, which require an in-depth understanding of both traditional computer science fundamentals and how to work effectively with AI tools.

Data Hub is an AI-native data platform developed by Datopian that allows users to discover, publish and interact with datasets through agentic workflows. It offers a curated selection of regularly maintained core datasets covering areas such as country codes, S&P 500 companies, airport codes and geographic boundaries, alongside enterprise-grade data solutions including postal codes, logistics data and country reference data.

For those unable to find what they need independently, a Premium Data Service connects users with data experts who will source, verify and integrate customised datasets on their behalf. The platform also supports easy publishing of data in a Markdown-based format with optional GitHub synchronisation and a free entry tier. A companion blog covers practical topics such as decoupling frontend architecture from CKAN-based portals and modernising open data infrastructure for organisations ranging from government bodies to private enterprises.

Claude has been integrated into Microsoft Excel as a beta feature available to Pro, Max, Team and Enterprise subscribers. The AI assistant can analyse entire workbooks, including complex formulas and dependencies across multiple tabs, whilst providing explanations with specific cell references for verification.

Users can test different scenarios by updating assumptions throughout their models without disrupting existing formulas, with all changes clearly highlighted and explained. The tool can also identify and help resolve common spreadsheet errors such as reference errors, value errors and circular dependencies by tracing them back to their origin.

Claude can generate draft financial models based on user requirements, and can populate existing templates with new data whilst preserving all formulas and structural elements. These capabilities make it a practical aid for both building and maintaining complex spreadsheet-based workflows.

Following user feedback that Claude Code was being applied to non-coding tasks, Anthropic has introduced Cowork, a simplified version designed for general productivity work rather than software development. Available initially as a research preview for Claude Max subscribers on macOS, Cowork allows users to grant Claude access to specific folders on their computers where it can read, edit and create files autonomously. The system can handle tasks such as reorganising downloads, generating spreadsheets from screenshots or drafting reports from notes, operating with greater independence than standard conversational interactions by making plans and executing them whilst keeping users informed of progress. Users can enhance Cowork's capabilities through existing connectors and newly added skills for document creation, and can combine it with Claude in Chrome for browser-based tasks.

Whilst users maintain control by selecting which folders Claude can access and receive prompts before significant actions occur, the system carries risks including potential file deletion through misinterpreted instructions and vulnerability to prompt injection attacks where malicious content might alter Claude's behaviour. The company plans to expand availability to other subscription tiers, add cross-device synchronisation and Windows support, and continue developing safety features based on feedback from this early release.

After open-weight language models made it cheaper and more practical to run capable systems outside proprietary platforms, many teams have found that hosting them locally still demands extreme hardware. Attention has therefore shifted to specialist API providers that charge by tokens and remove most of the infrastructure burden.

Several providers are compared using benchmark and live performance observations, with the assessment focusing on speed, latency, cost, accuracy and reliability. Cerebras stands out for very high throughput on large models, whilst Fireworks AI and Groq emphasise very low latency suited to interactive and real-time agent use. Together.ai aims for broadly strong, steady production performance on conventional GPU infrastructure, and Clarifai targets enterprise needs with hybrid deployment control and cost management. DeepInfra is presented as a lower-cost option suitable for batch or non-critical workloads, with the trade-off of weaker reliability compared to the leading services.

As software systems grow more complex and delivery cycles shorten, requirements engineering is under pressure to stay rigorous whilst moving faster. A recent article argues that artificial intelligence is increasingly being used to support that shift.

It outlines how AI can help teams capture and refine requirements earlier through meeting transcription and summarisation, automatic drafting of user stories and acceptance criteria, clustering and sentiment analysis to surface disagreement and themes, and live translation to improve collaboration in distributed teams. It also describes tools that generate diagrams, detect duplicates, suggest tests, support traceability and predict change impacts, alongside general-purpose assistants that help analysts brainstorm, rephrase and review specifications.

Alongside these potential gains in efficiency and consistency, the article stresses the need for careful governance around bias, privacy and transparency. Human oversight, clear accountability and compliance measures such as GDPR are described as remaining essential.

A recent study led by researchers at University College Cork used literary stylometry to compare hundreds of human-written short stories with creative prose generated by large language models such as GPT-3.5, GPT-4 and Llama 70B. The findings indicate that whilst AI can produce fluent and polished writing, it still shows a detectable stylistic fingerprint.

The analysis identified clear differences in subtle linguistic markers. AI outputs clustered into narrow, uniform patterns specific to each model, whereas human writing displayed far greater variation shaped by individual voice and creative intent. GPT-4 was found to be even more stylistically consistent than GPT-3.5.

The researchers caution that stylometry is not suitable for judging student authorship in education due to ethical and reliability concerns. They argue, however, that it can help clarify what distinguishes human expression from algorithmic generation, and they call for broader datasets and further testing as generative AI raises wider questions about authenticity and authorship in creative work.

CoCalc is a collaborative platform designed for computational work, offering real-time editing of Jupyter Notebooks, LaTeX documents and Linux terminals alongside access to powerful compute servers with optional GPU support. It provides pre-installed software such as Python, R, Julia and SageMath, along with tools for course management, automated grading and integration with generative AI models to assist with coding and documentation. Users can deploy projects on-premises or via cloud services, utilising scalable resources and flexible licensing options.

The platform supports extensive collaboration features, including shared workspaces, version control and the ability to synchronise files across multiple servers, making it suitable for academic, research and professional environments. Testimonials highlight its ease of use, robust support and effectiveness in facilitating both individual and group computational tasks.

JuliaHub offers two AI-powered platforms designed to modernise engineering and scientific computing workflows. Dyad combines traditional physics modelling with artificial intelligence to support digital twin development, system optimisation and embedded software deployment, whilst JuliaHub provides cloud-native infrastructure for high-performance scientific computing and AI workloads.

Both platforms are built on Julia, a programming language that delivers performance claimed to be fifty times faster than Python, MATLAB and R. The language has attracted over one million users, more than one hundred million downloads and upwards of twelve thousand packages across fifteen hundred universities. The company serves clients in aerospace, pharmaceutical and technology sectors, and publishes regular technical updates, product releases and case studies through its blog.