Resources

Find our analyses, experiences, and thoughts on tech, product, and organization here.

7/9/2026

Measuring AI success: beyond time saved

A case studied at the Community College of Philadelphia shows why saved hours are not enough to evaluate AI value. In some processes, the real gains appear in better decisions, fewer reworks, and clearer handoffs.

Neurotech: the real race is not only Neuralink

News & SignalsArtificial IntelligenceTech

6/25/2026

Neurotech: the real race is not only Neuralink

Brain-computer interfaces are moving beyond spectacular demos. The real question is becoming more concrete: surgery, reimbursement, patient access, clinical evidence, and the ability to move from lab result to care pathway.

AI, data centers, electricity: the new dependency chain for European companies

News & SignalsArtificial IntelligenceTech

6/25/2026

AI, data centers, electricity: the new dependency chain for European companies

Behind APIs and models, AI depends on data centers, electricity, grid connections, cloud regions, and model choices that also determine where data travel.

The next AI bottleneck for B2B teams is people, not tools

News & SignalsArtificial Intelligence

6/21/2026

The next AI bottleneck for B2B teams is people, not tools

SmarterX says 74% of respondents see AI as critical or very important to business success, yet the top adoption barriers are training, understanding, time, and trust. The useful reading for B2B teams is that AI advantage is becoming an operating-system problem, not just a tooling problem.

Artificial IntelligenceSociety

6/11/2026

When AI Turns Us Into Little Kittens

What if the real AI dystopia is not a war against machines, but a society where economic intelligence concentrates inside a few systems while many humans are fed, protected, and made less necessary?

Meta: when support AI becomes a security flaw

News & SignalsArtificial Intelligence

6/8/2026

Meta: when support AI becomes a security flaw

Meta’s Instagram account recovery incident shows a concrete AI operations risk: once an AI-assisted system touches account recovery, the danger is not only what it says, but what it can trigger.

Radiology AI: the real lever is workflow orchestration

News & SignalsArtificial IntelligenceTech

5/22/2026

Radiology AI: the real lever is workflow orchestration

In radiology, AI value does not sit only in diagnosis support, pathology detection, or reporting. The real lever is often operational: prioritizing, routing, contextualizing, and reviewing the right studies at the right time.

SaaStr AI traffic spike is a signal, not a market verdict

News & SignalsArtificial Intelligence

5/20/2026

SaaStr AI traffic spike is a signal, not a market verdict

SaaStr reports +96% active users, +114% new users, and +42% organic search across SaaStr.com and SaaStr.ai over a 28-day year-over-year snapshot. The useful reading is not that every AI content strategy works. It is that trusted B2B brands can still capture attention when they turn noisy AI demand into useful interpretation.

What GitHub’s accessibility agent reveals about operable coding agents

News & SignalsArtificial Intelligence

5/19/2026

What GitHub’s accessibility agent reveals about operable coding agents

GitHub’s post does not just show an agent that fixes code. It shows what makes an agent usable in real workflows: organizational memory, separated roles, guardrails, traceability, and human escalation.

I fixed a production bug from my phone with ChatGPT

Artificial IntelligenceJournal

4/27/2026

I fixed a production bug from my phone with ChatGPT

I had vibe-coded, with Google's Antigravity tool, a very small app to help my girlfriend explore the available job postings before the ESIRA ranking session. A few hours later, it stopped working. I was at physio, without my laptop. I opened ChatGPT on my phone, gave it access to the project, and the problem got resolved. What struck me most was how simple the whole thing was.

Two Years of Development with AI Tools: From ChatGPT in the Browser to Multi-Agent Orchestration (From 2024 to April 2026)

Artificial IntelligenceTech

4/18/2026

Two Years of Development with AI Tools: From ChatGPT in the Browser to Multi-Agent Orchestration (From 2024 to April 2026)

Two years ago, I started using ChatGPT 3.5 in a browser, next to my IDE. Today, I don't write code myself anymore. I orchestrate AI agents from my terminal, review their PRs, and manage a token budget the same way I used to manage a team budget.

When AI Codes: Assistants, Workflows, and Adoption in Software Engineering in 2025

Artificial IntelligenceTech

11/3/2025

When AI Codes: Assistants, Workflows, and Adoption in Software Engineering in 2025

The year 2025 marks a turning point in software development. AI is no longer just assisting developers—it's integrating into workflows, structuring production, and redefining the very role of the engineer. Analysis.

How to Restructure a Tech Team for More Agility and Velocity: The Team Topologies Approach in a Scale-up

OrganizationTech

10/9/2025

How to Restructure a Tech Team for More Agility and Velocity: The Team Topologies Approach in a Scale-up

Discover the principles of Team Topologies applied in a scale-up context to improve velocity and collaboration. Case study: the successful experience of Shift Technology.