Monzo's former CEO shares 3 tips for getting the most out of vibe coding
Former Monzo CEO Tom Blomfield shared tips on how to get the most out of vibe coding.
They include experimenting with different tools and keeping a log in case you need to reset to a clean code base.
Vibe coding continues to gain traction in Silicon Valley, and former Monzo CEO Tom Blomfield has thoughts on how to maximize its potential.
Coined just two months ago by Andrej Karpathy, an OpenAI cofounder, the term refers to people using AI to write code by giving it text-based instructions.
Experienced engineers are using it to save time, and those with nontechnical backgrounds are coding everything from dating apps to games.
Blomfield, who's now a group partner at Y Combinator, shared some tips for people looking to optimize the way they vibe code, in a video posted by the accelerator on Friday. Here are three pieces of advice he gave.
Pick the right tool and create a comprehensive plan
Blomfield advised users to plan ahead and experiment to find the tool that best supports their skill level and desired end product.
He found that tools like Lovable and Replit were suited for beginners, whereas more experienced coders could use Windsurf or Cursor.
"Work with the LLM to create a comprehensive plan," he said in the video, referring to large language models. "Put that in a markdown file inside your project folder and keep referring back to it."
He suggested that users could use the LLM to carry out the plan section by section, instead of making the product in one go.
"This advice could change in one or two months, as the models are getting better," he added.
Do version tests on the product
Blomfield said that when he prompted AI tools multiple times for the same coding task, he would get bad results as a result of the model accumulating "layers of bad code."
He advised using the large language model to write tests that simulate someone clicking through a version of the site or app, to gauge how well the features are working.
Sometimes, LLMs can make unnecessary changes to these features, he said, and implementing integration tests can pick up on these changes quicker.
Write instructions for the LLMs
Blomfield said he found that different models succeeded where others failed. If a user encounters a specific bug, it's helpful to reset all changes and give the LLM detailed instructions to fix it on a clean code base.
"Logging is your friend," Blomfield said.
Another tip he offered was to use small files and a more modular, service-based architecture, where the LLM has clear API boundaries.
An upside of this is that it would avoid creating a huge single repository of code for various projects, which could be more complex to manage and have more integration challenges.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
21 minutes ago
- Yahoo
CuspAI in Talks to Raise $100 Million to Discover New Materials
(Bloomberg) -- British startup CuspAI is in talks to raise more than $100 million in funding to support its goal of using artificial intelligence models to discover new materials, according to people familiar with the matter. PATH Train Service Resumes After Fire at Jersey City Station Seeking Relief From Heat and Smog, Cities Follow the Wind Chicago Curbs Hiring, Travel to Tackle $1 Billion Budget Hole NYC Mayor Adams Gives Bally's Bronx Casino Plan a Second Chance Founded in 2024, CuspAI uses generative AI and molecular simulation to build a platform that it likens to a highly specialized search engine. Users can describe properties they'd like a new material to have and the service responds with a chemical makeup. CuspAI declined to comment. The people familiar with the deal talks asked not to be identified discussing private information. CuspAI Chief Executive Officer Chad Edwards previously told Bloomberg News that he sees opportunity in green hydrogen, synthetic fuels and semiconductor manufacturing. The startup recently partnered with Kemira Oyj, a Finnish chemicals company, starting with a project focused on removing forever chemicals from water. The company raised $30 million in seed funding in 2024. Earlier this year, AI 'godfather' and recent Nobel laureate Geoffrey Hinton joined the startup's advisory board. AI Flight Pricing Can Push Travelers to the Limit of Their Ability to Pay How Podcast-Obsessed Tech Investors Made a New Media Industry Government Steps Up Campaign Against Business School Diversity What Happens to AI Startups When Their Founders Jump Ship for Big Tech Everyone Loves to Hate Wind Power. Scotland Found a Way to Make It Pay Off ©2025 Bloomberg L.P. Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
Yahoo
21 minutes ago
- Yahoo
Solving the unstructured data dilemma in financial services
I hear you. Not another highbrow article about AI. But stay with me. What follows isn't abstract speculation — it's rooted in what's already unfolding on the ground. In a recent conversation with Richard Huston, Managing Director at VAMOS, a company building AI solutions specifically for financial services, I heard something different. Rather than grand visions of the future, Huston offered a practical, behind-the-scenes look at how AI is quietly tackling one of the sector's most persistent pain points: unstructured data. His insights prompted the following. The unstructured data problem In the fast-paced world of financial services, few challenges are as persistent or as costly as the management of unstructured data. With thousands of finance proposals flooding in from hundreds of brokers to customer service emails requiring complex policy interpretations, financial institutions have long struggled with the manual processing of information that arrives in countless formats and styles. Yet, as in many areas, recent advancements in AI technology are offering promising solutions to this longstanding problem. The scale of the challenge is staggering. Consider the typical asset finance lender receiving proposals from brokers across the country. Each broker has their own approach to presenting client information; some prefer lengthy email narratives, others send sparse bullet points, while many submit incomplete documentation across multiple attachments. What should be a straightforward evaluation process becomes a laborious exercise in data archaeology, with operations teams spending hours extracting, interpreting and structuring information before any meaningful analysis can begin. The rule of 80:20 Emerging AI technologies are now capable of interpreting customer communications and providing intelligent suggestions to operations teams. Rather than promising to replace human judgment, these systems aim to enhance it by automating administrative heavy lifting: the 80 allowing human intelligence (HI), the 20 to focus on oversight and complex decision-making. Whether it's a broker's finance proposal or a customer service enquiry, when unstructured data arrives AI can automatically analyse the content, extract key information and present it in a structured format alongside relevant context from existing business systems. For example, when a customer emails to change their payment date, an AI system can analyse the request, fetch the customer's account information and payment schedule, and provide recommended actions based on policy and account status. Beyond simple automation What makes tech such as this particularly powerful is its ability to understand context and apply policy consistently across all interactions. In the previous example, the system doesn't just extract basic applicant information, it understands the relationship between different data points, identifies potential risk factors and suggests next steps based on established lending criteria. When a customer sends an email explaining that they are having financial difficulties, AI can analyse the email and understand the individual circumstances, retrieve relevant policies for customers in financial difficulty and offer staff appropriate options such as reduced payments or a payment holiday. This level of contextual understanding represents a significant leap forward from traditional data processing tools. Compliance and risk management In an industry where regulatory compliance is paramount, AI's approach to policy consistency provides significant value. It can ensure all customer interactions are automatically guided according to policies and procedures, while creating a clear audit trail of circumstances considered for taking decisions and enabling monitoring of customer outcomes. This automated compliance capability is particularly valuable for institutions dealing with Consumer Duty obligations and fair treatment requirements. By embedding policy guidance directly into the operational workflow, AI helps ensure that regulatory requirements are consistently met across all customer interactions, reducing compliance risk while improving operational efficiency. HI – the human factor Perhaps most importantly, successful AI implementation must account for the human element. It should reduce pressure on frontline staff by simplifying complex customer interactions, providing clear guidance and support while maintaining human oversight of key decisions and customer interactions. This human-centric approach addresses one of the most common concerns about AI in financial services – that automation will diminish the quality of customer service. Instead, by handling administrative tasks and providing intelligent suggestions, AI enables staff to spend more time on activities that truly require the nuance of human judgment and empathy. Looking forward AI's potential to solve the unstructured data challenge points to a broader transformation in how financial institutions approach operational efficiency. Rather than viewing AI as a replacement for human expertise, forward-thinking organisations are recognising its potential as a powerful augmentation tool that can handle routine processing while preserving human oversight for complex decisions. For financial institutions still struggling with manual processing of unstructured data, AI offers a compelling vision of the future – one where technology handles the administrative burden while humans focus on building relationships and driving business outcomes. In an industry where efficiency and compliance are both critical to success, this represents not just an operational improvement, but a genuine competitive advantage. The question for leaders in financial services isn't whether AI will transform their operations – it's whether they'll be early adopters of these solutions or find themselves playing catch-up in an increasingly automated industry. "Solving the unstructured data dilemma in financial services" was originally created and published by Motor Finance Online, a GlobalData owned brand. The information on this site has been included in good faith for general informational purposes only. It is not intended to amount to advice on which you should rely, and we give no representation, warranty or guarantee, whether express or implied as to its accuracy or completeness. You must obtain professional or specialist advice before taking, or refraining from, any action on the basis of the content on our site. 擷取數據時發生錯誤 登入存取你的投資組合 擷取數據時發生錯誤 擷取數據時發生錯誤 擷取數據時發生錯誤 擷取數據時發生錯誤
TechCrunch
22 minutes ago
- TechCrunch
Google says its AI-based bug hunter found 20 security vulnerabilities
Google's AI-powered bug hunter has just reported its first batch of security vulnerabilities. Heather Adkins, Google's vice president of security, announced Monday that its LLM-based vulnerability researcher Big Sleep found and reported 20 flaws in various popular open source software. Adkins said that Big Sleep, which is developed by the company's AI department DeepMind as well as its elite team of hackers Project Zero, reported its first-ever vulnerabilities, mostly in open source software such as audio and video library FFmpeg and image editing suite ImageMagick. Given that the vulnerabilities are not fixed yet, we don't have details of their impact or severity, as Google does not yet want to provide details, which is a standard policy when waiting for bugs to be fixed. But the simple fact that Big Sleep found these vulnerabilities is significant, as it shows these tools are starting to get real results, even if there was a human involved in this case. 'To ensure high quality and actionable reports, we have a human expert in the loop before reporting, but each vulnerability was found and reproduced by the AI agent without human intervention,' Google's spokesperson Kimberly Samra told TechCrunch. Royal Hansen, Google's vice president of engineering, wrote on X that the findings demonstrate 'a new frontier in automated vulnerability discovery.' LLM-powered tools that can look for and find vulnerabilities are already a reality. Other than Big Sleep, there's RunSybil, and XBOW, among others. Techcrunch event Tech and VC heavyweights join the Disrupt 2025 agenda Netflix, ElevenLabs, Wayve, Sequoia Capital — just a few of the heavy hitters joining the Disrupt 2025 agenda. They're here to deliver the insights that fuel startup growth and sharpen your edge. Don't miss the 20th anniversary of TechCrunch Disrupt, and a chance to learn from the top voices in tech — grab your ticket now and save up to $675 before prices rise. Tech and VC heavyweights join the Disrupt 2025 agenda Netflix, ElevenLabs, Wayve, Sequoia Capital — just a few of the heavy hitters joining the Disrupt 2025 agenda. They're here to deliver the insights that fuel startup growth and sharpen your edge. Don't miss the 20th anniversary of TechCrunch Disrupt, and a chance to learn from the top voices in tech — grab your ticket now and save up to $675 before prices rise. San Francisco | REGISTER NOW XBOW has garnered headlines after it reached the top of one of the U.S. leaderboards at bug bounty platform HackerOne. It's important to note that in most cases, these reports have a human at some point of the process to verify that the AI-powered bug hunter found a legitimate vulnerability, as is the case with Big Sleep. Vlad Ionescu, co-founder and chief technology officer at RunSybil, a startup that develops AI-powered bug hunters, told TechCrunch that Big Sleep is a 'legit' project, given that it has 'good design, people behind it know what they're doing, Project Zero has the bug finding experience and DeepMind has the firepower and tokens to throw at it.' There is obviously a lot of promise with these tools, but also significant downsides. Several people who maintain different software projects have complained of bug reports that are actually hallucinations, with some calling them the bug bounty equivalent of AI slop. 'That's the problem people are running into, is we're getting a lot of stuff that looks like gold, but it's actually just crap,' Ionescu previously told TechCrunch.
