Human Vision + AI Assistance
I've been taking advantage of Gen AI to help me rebuild an old project
This week, I'm going to bring you some insight into a different project I have been working on. Many years ago, I built a website (from scratch with Ruby on Rails) that focuses on Coffee. Local coffee shops and coffee roasters. Ultimately I wanted to allow users to indicate their preferences for specific coffee roasts and have the website make recommendations. I failed with that. It was technically beyond me as a very, very basic coder. However, with the raise of generative AI, I have picked this project back up, starting from scratch with an entirely new coding language I don't know.
My AI Coding Assistants
Within a week, I've created a fairly sophisticated solution that almost fully replicates my old legacy site. I could never have done this without using AI, with my tools of choice being Google Gemini and Perplexity. For reference, my old site took me nearly a full year, to get it to where it is today. With a ton of trial and error and asking many, many questions on Stackoverflow. The speed I've been able to develop at, in a coding language I don't know at all, is incredible.
Human Vision - AI Assist
However, there is an important aspect to my workflow that I would like to highlight. I have the overall vision and goal of what I want to achieve with my website. How it looks, how it operates and what features it offers. I am using LLMs to purely help me write the code. This is different from what you might have heard called 'Vibe Coding'. This is where you prompt an LLM with what you want to create, and it does it all for you. There are specific tools out there like, Replit, Bolt and Lovable, that will build a complete website from a single prompt. These are great of course, but you lack some of the control and details of what's going on.
In my workflow I am far more intentional about asking Gemini to create me a navigation bar, or to add a page footer and this is a perfect blend of my human creativity coupled with AI technical capability.
Where Could You Use This
You might not want to build a website about coffee, but do you have some manual process that could potentially be automated? Where can you take advantage of this "vision + AI co-pilot"?
- Maybe you have data analysis tasks they can't do themselves – AI can help write scripts or explain complex functions.
- Maybe you need to automate a process but don't know where to start – AI can suggest steps or even generate low-code workflow structures.
Have a think, and let me know if you implement anything.
Andrew Filev, CEO and founder of Zencoder
This week's email is brought to you in partnership with Capsule.
Too many CRMs feel like they’re built for massive sales teams. Capsule is different. It’s built for small businesses that want a simple, effective way to track customers and sales, without spending hours figuring it out.
- Easy to use
- Helps you close more deals
- No unnecessary features
Curated News
Agentic AI Startups Secure $36 Million in Funding
In April 2025, four AI startups, Qevlar, HoneyHive, Solve Intelligence and Delos, collectively raised nearly $36 million from prominent investors. Qevlar focuses on AI-powered cybersecurity incident investigation, while HoneyHive develops platforms for AI agent testing and evaluation. Solve Intelligence offers AI-driven tools for intellectual property patenting, and Delos works on generative AI for office productivity tasks. These investments reflect growing confidence in agentic AI applications across various industries.
Nvidia's NeMo Microservices: Empowering Custom AI Agents
Nvidia has unveiled NeMo microservices, a software platform enabling businesses to develop autonomous AI agents using open-weight models from sources like Meta and Mistral AI. This approach offers enterprises greater flexibility and control compared to proprietary models. NeMo aims to simplify the integration of private data into AI agent development, addressing challenges in training models with confidential business information. By providing this platform, Nvidia seeks to strengthen its position in the AI market, estimating the AI agent market to be worth $1 trillion.
AI-Powered Smart Cities: Melbourne's Innovative Approach
A collaborative report by Deloitte, ServiceNow, Nvidia, and ThoughtLab highlights the integration of agentic AI in urban management. Melbourne stands out with its use of AI in real-time water quality forecasting and monitoring illegal waste dumping. These applications demonstrate how AI can enhance decision-making and automate routine processes in city operations. The report underscores the importance of strategic leadership, modern IT infrastructure, and cross-sector collaboration for successful AI integration in urban settings.
Upcoming AI Events
Rise of AI Conference
Berlin, Germany, May 7–8
AI in Business Conference
London, May 15
Google I/O
Shoreline Amphitheatre, Mountain View, CA, USA, June 9–12
Thanks for reading, and see you next Friday.
Simon,
Are you looking for a job in the tech sector? You'd do well to check out all the available roles on Wellfound. Wellfound is a platform that connects startups and tech companies with job seekers. It offers over 130,000 remote and local job listings, allowing candidates to apply with a single profile and view salary and equity details upfront.
You can also follow me on social media:
Comments ()