AI

10 posts
AI overdose: When developers stop thinking and start prompting

AI overdose: When developers stop thinking and start prompting

AI tools are reshaping how junior engineers approach problems, often replacing simple solutions with overly complex ones. Here’s why foundational thinking still matters. A real-life case of AI over-engineering gone wrong highlights why understanding problem domains still beats prompting.

Tech stack decisions for AI startups: what you need to know

Tech stack decisions for AI startups: what you need to know

Selecting the right tech stack is critical for AI startups. Python is essential for data science, but the backend, frontend, and infrastructure choices determine scalability and efficiency. Explore the best tech stack combinations, hosting tools, and ETL solutions to future-proof your AI startup.

Exploring LLM Options: Proprietary vs. Open Source

Exploring LLM Options: Proprietary vs. Open Source

AI can be challenging. We break down three AI integration tiers—proprietary models, open-source solutions, and custom-built systems—to help you choose the right approach. From quick MVPs to scalable solutions, discover how to leverage AI effectively for your product.

Building the dream team for an AI startup

Building the dream team for an AI startup

Building an AI startup demands specialised roles like data scientists, engineers, and analysts to drive innovation. Discover what roles you need to hire first for a strong foundation to success.

Even ETL needs a roadmap

Even ETL needs a roadmap

ETL (Extract, Transform, Load) is vital for data-heavy businesses but often begins with manual workflows. Companies should identify inefficiencies, prioritise automation, and design a scalable ETL roadmap that integrates human reviews and evolves with business growth.

Revolutionising diagnosis: How large language models can drive a 10x change

Revolutionising diagnosis: How large language models can drive a 10x change

Large language models (LLMs) transform problem-solving by enabling natural, iterative conversations, ideal for fields like healthcare and legal services. They scale expertise and accessibility but face challenges like reliability and cost.

Do we even need a moat?

Do we even need a moat?

A technical moat is often seen as a product's defensive edge, but does every product really need one? For AI products, the choice between building proprietary tech or leveraging existing solutions like OpenAI is complex. True value lies in solving customer problems—not just in owning the technology.

A company's first steps in AI

A company's first steps in AI

Many product companies are eager to leverage tools like ChatGPT. But how do you go from experimenting to running in production? Let's explore choosing the right large language model (LLM) to understand hosting options, ensuring an efficient and sustainable AI implementation.

The fate of being too early

The fate of being too early

Strava's new AI-driven Athlete Intelligence brought me back to Addapp, the startup I co-founded nearly a decade ago. We used data from devices like Apple Watch to offer insights, like how your run affected your sleep, using traditional data science. Being ahead of time is not always good.

You’ve successfully subscribed to madewithlove
Welcome back! You’ve successfully signed in.
Great! You’ve successfully signed up.
Success! Your email is updated.
Your link has expired
Success! Check your email for magic link to sign-in.