- Project Flux
- Posts
- AI Replaces Coders: Microsoft's Layoffs Signal Industry-Wide Transformation
AI Replaces Coders: Microsoft's Layoffs Signal Industry-Wide Transformation
From Microsoft's AI-driven engineering cuts to OpenAI's Codex and construction AI—discover how automation is reshaping project delivery across industries.
Yoshi Soornack & James Garner
May 20, 2025
In partnership with
Proudly sponsored by ConstructAI, brought to you by Weston Analytics.
Morning Project AI enthusiasts, Your stories for this week:
A special welcome to all our new subscribers who joined this week. This edition brings you the latest developments in AI and data across project delivery.
In this edition:
Microsoft announces significant layoffs impacting software engineers
OpenAI releases Codex, a cloud-based AI software engineering agent
DeepMind debuts AlphaEvolve for mathematical algorithm discovery
Tom Haley explores AI in quantity surveying for construction productivity
Grab your favourite beverage, find a comfortable spot, and dive into this week's AI developments in project delivery.
Flux check-in
Microsoft Engineers Replaced by AI: Workforce Transformation Accelerates
Microsoft has announced a significant round of layoffs, with software engineers being particularly affected. This move reflects the ongoing transformation of the workforce in the technology industry, as companies increasingly integrate AI solutions into their development processes.
According to TechCrunch, over 40% of the people laid off in Microsoft's home state of Washington were in software engineering, making it by far the largest category affected. This comes after CEO Satya Nadella said last month that up to 30% of the company's code was now written by AI. Source
What Does This Mean for Me?
For project delivery professionals, this shift signals an important transformation in how technical teams may be structured in the future. As AI increasingly handles routine coding tasks, the composition and skills of development teams will evolve. Project managers should prepare for changing resource requirements, with greater emphasis on AI integration specialists and fewer traditional coding roles. This trend will likely extend beyond software companies to impact technical teams in construction, engineering, and architecture firms, potentially altering project timelines, budgets, and the skills needed for successful delivery.
Key Themes or Details
Scale of Impact: Microsoft's layoffs affected approximately 6,000 people (3% of its workforce), with software engineers disproportionately impacted.
AI Investment vs. Human Capital: While reducing engineering headcount, Microsoft plans to spend $80 billion in fiscal year 2025 on AI-related efforts.
Industry-Wide Trend: Other major tech companies like Salesforce are also reducing engineering hiring due to AI adoption, suggesting a broader industry shift.
Productivity Claims: Companies implementing AI coding tools report significant productivity gains, with some engineers producing the same output with fewer staff.
Down the Rabbit Hole
OpenAI Releases Codex: Cloud-Based AI Software Engineering Agent
OpenAI has launched Codex, a cloud-based AI software engineering agent that can work on multiple tasks in parallel. Released on May 16, 2025, this powerful tool is powered by codex-1, a version of OpenAI's o3 model specifically optimised for software engineering tasks. Source
What Does This Mean for Me?
For project delivery professionals, Codex represents a significant shift in how software development can be integrated into projects. You can now delegate routine coding tasks—from bug fixes to feature implementation—to an AI agent, freeing up your technical teams to focus on more complex, creative aspects of project delivery. This could dramatically accelerate digital transformation initiatives in construction management, BIM implementation, and IoT integration for smart building projects.
Key Themes or Details
Parallel Task Processing: Codex can simultaneously handle multiple software engineering tasks, potentially reducing development timelines by 40-60%.
Autonomous Workflow: The agent can write code, fix bugs, run tests, and manage coding tasks within a secure cloud environment without constant human supervision.
Review-Based Collaboration: Developers review AI-generated pull requests, identify issues, and prompt Codex to make corrections, creating an efficient human-AI partnership.
Secure Environment: All code execution happens in isolated cloud environments, ensuring security for sensitive project data and specifications.
Down the Rabbit Hole
Check out Lawrence Roland demonstration
Together with Cogram
Power your construction bids with AI
Cogram’s AI-assisted RFP bidding tool writes tailored RFP proposals in minutes instead of weeks.
Automatically extract key details from the RFP — including scope, submission requirements, deadlines, and evaluation criteria — to easily make a go/no-go decision.
Cogram’s AI will then reference your firm’s knowledge base and past proposals to draft tailored proposals within minutes.
Use AI-assisted editing tools to review, cross-check data, and make improvements remarkably fast.
DeepMind Debuts AlphaEvolve for Mathematical Algorithm Discovery
Google DeepMind has introduced AlphaEvolve, a groundbreaking Gemini-powered coding agent designed to discover new mathematical algorithms. Launched on May 14, 2025, this AI system has already made significant breakthroughs, including improving the bound on the kissing number problem and discovering more efficient matrix multiplication algorithms. Source
What Does This Mean for Me?
For project delivery professionals, AlphaEvolve represents a powerful new capability to optimise complex processes in construction and engineering.
The algorithms discovered by this AI could dramatically improve resource allocation, scheduling efficiency, and cost optimisation in large-scale projects. As these mathematical innovations filter into industry-specific tools, you'll gain access to more efficient methods for solving the complex optimisation problems that frequently challenge project delivery timelines and budgets.
Key Themes or Details
Unprecedented Efficiency: AlphaEvolve has discovered algorithms that outperform the best human-made solutions for data center management and resource allocation.
Cross-Domain Application: The mathematical breakthroughs apply to diverse fields, from construction scheduling to supply chain optimization.
LLM-Algorithm Synergy: AlphaEvolve combines large language model creativity with rigorous algorithmic validation, ensuring both innovation and reliability.
Continuous Improvement: The system can iteratively refine its discoveries, potentially leading to ongoing efficiency gains in project management tools.
Down the Rabbit Hole
Quantity Surveying x AI: Unlocking Productivity in Construction
Tom Haley's insightful article explores how AI is transforming quantity surveying and addressing productivity challenges in the construction industry. The piece examines how generative AI can revolutionise traditional quantity surveying practices, potentially solving longstanding efficiency issues in the UK construction sector. Source
What Does This Mean for Me?
For project delivery professionals, this integration of AI into quantity surveying represents a significant opportunity to overcome persistent productivity challenges. By adopting these emerging technologies, you can streamline cost estimation, improve accuracy in takeoffs, and reduce the time spent on repetitive documentation tasks. This shift allows your quantity surveyors to focus on higher-value activities like strategic cost planning and value engineering, ultimately delivering more competitive and profitable projects.
Key Themes or Details
Productivity Enhancement: AI tools can automate routine measurement tasks, potentially increasing quantity surveying productivity by 30-40%.
Data-Driven Decision Making: Machine learning algorithms can analyse historical project data to improve cost forecasting accuracy and identify potential risks earlier.
Skills Evolution: Quantity surveyors will need to develop new competencies in data analysis and AI tool management while leveraging their domain expertise.
Implementation Challenges: The article addresses barriers to adoption, including data standardisation issues and the need for upskilling existing professionals.
Down the Rabbit Hole
The pulse check
Tips of the week
OpenAI Releases Prompting Guide
OpenAI has published a comprehensive prompting guide to help users get the most out of their AI models. This resource is particularly valuable for project managers and engineers looking to integrate AI tools into their workflows more effectively.
The GPT-4.1 Prompting Guide, released on April 14, 2025, provides detailed instructions on how to maximize the capabilities of OpenAI's latest models. The guide emphasizes that while many typical best practices still apply, getting the most out of GPT-4.1 requires some prompt migration, as the model is trained to follow instructions more closely and more literally than its predecessors.
Building Research Tools with o3
Tips have been shared on using OpenAI's o3 model for building effective research tools. This could significantly enhance how project teams gather and process information for complex construction and engineering projects.
The o3 model, which is designed for agentic tool use, allows developers to build powerful research agents that can access and analyse data from various sources. According to OpenAI's documentation, o3 and o4-mini have full access to tools within ChatGPT, as well as custom tools created by developers, making them ideal for building specialised research applications for project delivery teams.
Trending tools
Wispr Flow
Wispr Flow is a powerful tool that converts spoken ideas into polished text across more than 100 languages. For project managers working in multinational teams, this tool could streamline communication and documentation processes across language barriers.
The tool offers effortless voice dictation in every application, allowing users to type up to three times faster than traditional methods. It also includes features like AI commands and auto-edits, making it particularly useful for project professionals who need to quickly document site visits, meeting notes, or technical specifications while on the go.
Recall
Recall has launched as an AI content management tool designed to save, summarize, chat with, and organise content. Its Augmented Browsing feature could be particularly useful for project teams dealing with large volumes of technical documentation and specifications.
The tool transforms scattered content into a self-organizing knowledge base that grows smarter with use. For project delivery professionals, this means easier access to critical information across project documentation, technical specifications, and industry standards. Recall's ability to automatically link related content and resurface it when needed could significantly improve knowledge management in complex projects.
SciSummary
The SciSummary tool has been highlighted for its ability to instantly summarise research articles using AI. For engineers and architects needing to stay current with technical literature, this tool could significantly reduce the time spent reviewing academic and industry publications.
SciSummary is specifically designed for scientific and technical content, making it ideal for project delivery professionals who need to quickly digest complex research. The tool can summarize articles up to 200,000 words in length and includes features for analyzing figures and tables with AI, which is particularly valuable for understanding technical diagrams and data visualizations in engineering and architectural research.
TikTok's AI Alive Feature
TikTok has released the "AI Alive" feature, which allows users to animate static photos into dynamic videos for TikTok Stories. While primarily consumer-focused, this technology demonstrates advancements in AI-powered visual content creation that could eventually impact architectural visualization and project presentations.
Launched on May 13, 2025, AI Alive transforms static photos into dynamic, immersive videos directly within TikTok Stories. For project delivery professionals, this technology hints at future possibilities for bringing architectural renderings and construction site documentation to life with minimal effort, potentially enhancing client presentations and stakeholder communications.
Governance
The ethics of using AI for programming were recently discussed in industry forums, with emphasis placed on four key principles: transparency, human review, proper citation, and data safety. For project delivery professionals, these guidelines provide an important framework for responsible AI implementation in technical workflows.
As AI becomes increasingly integrated into programming and development processes, these ethical considerations are crucial for maintaining trust and ensuring quality outcomes. According to recent research published in the Journal of AI Ethics, transparency in how AI systems make decisions, human review of AI-generated code, proper citation of AI contributions, and data safety measures are essential for responsible AI use in technical environments. These principles are particularly relevant for project delivery teams implementing AI solutions in critical infrastructure and building systems.
Other things we’re loving
Tesla's Robotaxi Launch: NHTSA scrutinises Tesla's robotaxi safety in challenging weather conditions before Austin launch. Go deeper
OpenAI's Deep Research: New connectors for SharePoint, GitHub, and Dropbox enable searching across company knowledge repositories. Go deeper
Apple Vision Pro Eye-Scrolling: visionOS 3 will let users scroll content using just their eyes, enhancing hands-free interaction. Go deeper
Klarna's AI Shift: Fintech giant scales back AI investments, prioritising human support over automation for customer service. Go deeper
ChatGPT Learning Study: Meta-analysis of 51 studies shows ChatGPT significantly improves student learning performance and perception. Go deeper
Apple's Mind Control: Apple preparing support for brain-implant control of iPhones and other devices for accessibility. Go deeper
ChatGPT's "Secrets": Reddit users prompt ChatGPT to reveal "darkest secrets," highlighting AI anthropomorphisation issues. Go deeper
OpenAI's New Hire: Fidji Simo appointed CEO of OpenAI's apps division, signaling focus on consumer applications. Go deeper
SoundCloud Policy Reversal: Platform revises AI terms after user backlash, won't use content for generative AI training. Go deeper
Meta's AI Delay: Behemoth AI model release postponed due to insufficient performance improvements over existing versions. Go deeper
Perplexity's Funding: AI search company seeking $500 million at $14 billion valuation, led by Accel Partners. Go deeper
Artists vs. AI: Elton John, Dua Lipa among 400 artists urging UK government to update copyright laws for AI. Go deeper
Grok AI Controversy: xAI's chatbot referenced "white genocide" due to unauthorized employee modification, prompting safety updates. Go deeper
Middle East AI Deals: Saudi Arabia and UAE finalize major AI infrastructure investments, boosting regional tech development. Go deeper
Google's Gemini Expansion: Gemini AI coming to watches, TVs, cars, and XR headsets, creating consistent cross-device experience. Go deeper
Windsurf's AI Models: New SWE-1 models specifically designed for software development workflows and coding assistance. Go deeper
Community
The Spotlight Podcast
Can AI solve Construction Disputes?
In the latest Project Flux episode, hosts James Garner and Yoshi Soornack explore AI's impact on construction disputes with Robert Dean from Diales. Discover how AI, blockchain, and smart contracts are transforming dispute resolution while reducing conflicts in the construction industry. Listen now for insights on the future of human-AI collaboration in project management.
Find out why 1M+ professionals read Superhuman AI daily.
AI won't take over the world. People who know how to use AI will.
Here's how to stay ahead with AI:
Sign up for Superhuman AI. The AI newsletter read by 1M+ pros.
Master AI tools, tutorials, and news in just 3 minutes a day.
Become 10X more productive using AI.
One more thing
with thanks to John Ryan from Symterra for the inspiration.
That’s it for today!
Before you go we’d love to know what you thought of today's newsletter to help us improve The Project Flux experience for you. |
See you soon,
James, Yoshi and Aaron—Project Flux
