Latest news with #TobiLutke
Finextra
06-07-2025
- Business
- Finextra
Context Engineering for Financial Services: By Steve Wilcockson
The hottest discussion in AI right now, at least the one not about Agentic AI, is about how "context engineering" is more important than prompt engineering, how you give AI the data and information it needs to make decisions, and it cannot (and must not) be a solely technical function. "'Context' is actually how your company operates; the ideal versions of your reports, documents & processes that the AI can use as a model; the tone & voice of your organization. It is a cross-functional problem.' So says renowned Tech Influencer and Associate Professor at Wharton School, Ethan Molick. He in turn cites fellow Tech Influencer Andrej Karpathy on X, who in turn cites Tobi Lutke, CEO of Shopify: "It describes the core skill better: the art of providing all the context for the task to be plausibly solvable by the LLM. " The three together - Molick, Karpathy and Lutke - make for a powerful triumvirate of Tech-influencers. Karpathy consolidates the subject nicely. He emphasizes that in real-world, industrial-strength LLM applications, the challenge entails filling the model's context window with just the right mix of information. He thinks about context engineering as both a science—because it involves structured systems and system-level thinking, data pipelines, and optimization —and an art, because it requires intuition about how LLMs interpret and prioritize information. His analysis reflects two of my predictions for 2025 one highlighting the increasing impact of uncertainty and another a growing appreciation of knowledge. Tech mortals offered further useful comments on the threads, two of my favorites being: 'Owning knowledge no longer sets anyone apart; what matters is pattern literacy—the ability to frame a goal, spot exactly what you don't know, and pull in just the right strands of information while an AI loom weaves those strands into coherent solutions.' weaves those strands into coherent solutions.' 'It also feels like 'leadership' Tobi. How to give enough information, goal and then empower.' I love the AI loom analogy, in part because it corresponds with one of my favorite data descriptors, the "Contextual Fabric". I like the leadership positivity too, because the AI looms and contextual fabrics, are led by and empowered by humanity. Here's my spin, to take or leave. Knowledge, based on data, isn't singular, it's contingent, contextual. Knowledge and thus the contextual fabric of data on which it is embedded is ever changing, constantly shifting, dependent on situations and needs. I believe knowledge is shaped by who speaks, who listens, and what about. That is, to a large extent, led by power and the powerful. Whether in Latin, science, religious education, finance and now AI, what counts as 'truth' is often a function of who gets to tell the story. It's not just about what you know, but how, why, and where you know it, and who told you it. But of course it's not that simple; agency matters - the peasant can become an abbot, the council house schoolgirl can become a Nobel prize-winning scientist, a frontier barbarian can become a Roman emperor. For AI, truth to power is held by the big tech firms and grounded on bias, but on the other it's democratizing in that all of us and our experiences help train and ground AI, in theory at least. I digress. For AI-informed decision intelligence, context will likely be the new computation that makes GenAI tooling more useful than simply being an oft-hallucinating stochastic parrot, while enhancing traditional AI - predictive machine learning, for example - to be increasingly relevant and affordable for the enterprise. Context Engineering for FinTech Context engineering—the art of shaping the data, metadata, and relationships that feed AI—may become the most critical discipline in tech. This is like gold for those of us in the FinTech data engineering space, because we're the dudes helping you create your own context. I'll explore how five different contextual approaches, all representing data engineering-relevant vendors I have worked for —technical computing, vector-based, time-series, graph and geospatial platforms—can support context engineering. Parameterizing with Technical Computing Technical computing tools – think R, Julia, MATLAB and Python's SciPy stack - can integrate domain-specific data directly into the model's environment through structured inputs, simulations, and real-time sensor data, normally as vectors, tables or matrices. For example, in engineering or robotics applications, an AI model can be fed with contextual information such as system dynamics, environmental parameters, or control constraints. Thus the model can make decisions that are not just statistically sound but also physically meaningful within the modeled system. They can dynamically update the context window of an AI model, for example in scenarios like predictive maintenance or adaptive control, where AI must continuously adapt to new data. By embedding contextual cues, like historical trends, operational thresholds, or user-defined rules, such tools help ground the model's outputs in the specific realities of the task or domain. Financial Services Use Cases Quantitative Strategy Simulation Simulate trading strategies and feed results into an LLM for interpretation or optimization. Stress Testing Financial Models Run Monte Carlo simulations or scenario analyses and use the outputs to inform LLMs about potential systemic risks. Vectors and the Semantics of Similarity Vector embeddings are closely related to the linear algebra of technical computing, but they bring semantic context to the table. Typically stored in so-called vector databases, they encode meaning into high-dimensional space, allowing AI to retrieve through search not just exact matches, but conceptual neighbors. They thus allow for multiple stochastically arranged answers, not just one. Until recently, vector embeddings and vector databases have been primary providers of enterprise context to LLMs, shoehorning all types of data as searchable mathematical vectors. Their downside is their brute force and compute-intensive approach to storing and searching data. That said, they use similar transfer learning approaches – and deep neural nets – to those that drive LLMs. As expensive, powerful brute force vehicles of Retrieval-Augmented Generation (RAG), vector databases don't simply just store documents but understand them, and have an increasingly proven place for enabling LLMs to ground their outputs in relevant, contextualized knowledge. Financial Services Use Cases Customer Support Automation Retrieve similar past queries, regulatory documents, or product FAQs to inform LLM responses in real-time. Fraud Pattern Matching Embed transaction descriptions and retrieve similar fraud cases to help the model assess risk or flag suspicious behavior. Time-Series, Temporal and Streaming Context Time-series database and analytics providers, and in-memory and columnar databases that can organize their data structures by time, specialize in knowing about the when. They can ensure temporal context—the heartbeat of many use cases in financial markets as well as IoT, and edge computing- grounds AI at the right time with time-denominated sequential accuracy. Streaming systems, like Kafka, Flink, et al can also facilitate the real-time central nervous systems of financial event-based systems. It's not just about having access to time-stamped data, but analyzing it in motion, enabling AI to detect patterns, anomalies, and causality, as close as possible to real time. In context engineering, this is gold. Whether it's fraud that happens in milliseconds or sensor data populating insurance telematics, temporal granularity can be the difference between insight and noise, with context stored and delivered by what some might see as a data timehouse. Financial Services Use Cases Market Anomaly Detection Injecting real-time price, volume, and volatility data into an LLM's context allows it to detect and explain unusual market behavior. High-Frequency Trading Insights Feed LLMs with microsecond-level trade data to analyze execution quality or latency arbitrage. Graphs That Know Who's Who Graph and relationship-focussed providers play a powerful role in context engineering by structuring and surfacing relationships between entities that are otherwise hidden in raw data. In the context of large language models (LLMs), graph platforms can dynamically populate the model's context window with relevant, interconnected knowledge—such as relationships between people, organizations, events, or transactions. They enable the model to reason more effectively, disambiguate entities, and generate responses that are grounded in a rich, structured understanding of the domain. Graphs can act as a contextual memory layer through GraphRAG and Contextual RAG, ensuring that the LLM operates with awareness of the most relevant and trustworthy information. For example, graph databases - or other environments, e.g. Spark, that can store graph data types as accessible files, e.g. Parquet, HDFS - can be used to retrieve a subgraph of relevant nodes and edges based on a user query, which can then be serialized into natural language or structured prompts for the LLM. Platforms that focus graph context around entity resolution and contextual decision intelligence can enrich the model's context with high-confidence, real-world connections—especially useful in domains like fraud detection, anti-money laundering, or customer intelligence. Think of them as like Shakespeare's Comedy of Errors meets Netflix's Department Q. Two Antipholuses and two Dromios rather than 1 of each in Comedy of Errors? Only 1 Jennings brother to investigate in Department Q's case, and where does Kelly MacDonald fit into anything? Entity resolution and graph context can help resolve and connect them in a way that more standard data repositories and analytics tools struggle with. LLMs cannot function without correct and contingent knowledge of people, places, things and the relationships between them, though to be sure many types of AI can also help discover the connections and resolve entities in the first place. Financial Services Use Cases AML and KYC Investigations Surface hidden connections between accounts, transactions, and entities to inform LLMs during risk assessments. Credit Risk Analysis Use relationship graphs to understand borrower affiliations, guarantors, and exposure networks. Seeing the World in Geospatial Layers Geospatial platforms support context engineering by embedding spatial awareness into AI systems, enabling them to reason about location, proximity, movement, and environmental context. They can provide rich, structured data layers (e.g., terrain, infrastructure, demographics, weather) that can be dynamically retrieved and injected into an LLM's context window. This allows the model to generate responses that are not only linguistically coherent but also geographically grounded. For example, in disaster response, a geospatial platform can provide real-time satellite imagery, flood zones, and population density maps. This data can be translated into structured prompts or visual inputs for an AI model tasked with coordinating relief efforts or summarizing risk. Similarly, in urban planning or logistics, geospatial context helps the model understand constraints like traffic patterns, zoning laws, or accessibility. In essence, geospatial platforms act as a spatial memory layer, enriching the model's understanding of the physical world and enabling more accurate, context-aware decision-making. Financial Services Use Cases Branch Network Optimization Combine demographic, economic, and competitor data to help LLMs recommend new branch locations. Climate Risk Assessment Integrate flood zones, wildfire risk, or urban heat maps to evaluate the environmental exposure of mortgage and insurance portfolios. Context Engineering Beyond the Limits of Data, Knowledge & Truths Context engineering I believe recognizes that data is partial, and that knowledge and perhaps truth or truths needs to be situated, connected, and interpreted. Whether through graphs, time-series, vectors, tech computing platforms, or geospatial layering, AI depends on weaving the right contextual strands together. Where AI represents the loom, the five types of platforms I describe are like the spindles, needles, and dyes drawing on their respective contextual fabrics of ever changing data, driving threads of knowledge—contingent, contextual, and ready for action.
Crypto Insight
14-06-2025
- Business
- Crypto Insight
Shopify launches early access to USDC stablecoin payments on Base
Global e-commerce giant Shopify is rolling out early access to stablecoin payments in Circle's USDC in collaboration with major US exchange Coinbase. Shopify plans to fully roll out USDC payments on Coinbase's Ethereum layer-2 (L2) network Base via Shopify Payments and Shop Pay later this year, the company announced on Thursday. As part of the early access rollout, a limited number of merchants immediately have access to the full product, starting on June 13, a spokesperson for Shopify told Cointelegraph. 'We think that stablecoins are a natural way to transact on the internet and worked with Coinbase to develop the commerce payment protocol smart contract that powers this work,' Shopify CEO Tobi Lutke said in an X post on Thursday. The new stablecoin payment feature by Shopify will also allow the company to offer buyer incentives like 1% cash back in local currency payouts in the future, the CEO noted. The Shopify spokesperson said the new launch was both a technical build and a strategic partnership with the Base team. 'Buyers will need a wallet to pay with USDC. They can choose from hundreds of eligible wallets,' the representative added. Shopify taps Base for payments Coinbase's Base blockchain is the fourth-largest network for USDC, with Base-issued USDC accounting for 6% of the stablecoin's total supply of $61 billion, according to data from USDC Transparency and CoinGecko. Built by Coinbase, Base is an 'ultra-fast and affordable network that has emerged as a great way for moving money,' Shopify said. The company also mentioned that Base offers 'fast, cheap, and secure transactions' while providing a 24/7 global payment rail. The top five blockchains for USDC by supply at the time of writing. Source: USDC Transparency, CoinGecko Since Shopify did not mention whether the company expects to consider the native support of more USDC chains or just some additional crypto assets and stablecoins, many online commentators were curious about the choice of Base. 'What's the point of narrowing your top of the funnel? You should support all chains that stripe via USDC supports,' one user wrote on X. Indirect Bitcoin support in place since 2013 Shopify's new USDC partnership with Coinbase is not its first endeavor into crypto. Shopify has been indirectly supporting Bitcoin payments through gateway integrations since at least 2013, when the company officially announced that all of its 75,000 merchants were free to start accepting Bitcoin. According to Shopify Help Center, Shopify allows merchants to integrate at least nine additional payment methods featuring a wide variety of supported crypto assets, through integrations with third-party gateways like BitPay, Solana Pay and more. 'Due to longer settlement times, cryptocurrency transactions can cause overselling in flash sales. Use a direct payment method like Shopify Payments for the best flash sale performance,' the help center's message notes. The latest partnership between Shopify and Coinbase is not the first collaboration between the two companies using digital currency. In 2019 and 2020, Coinbase and Shopify joined Meta's (formerly Facebook's) stablecoin project Diem, initially known as Libra. Following years of pushback from global regulators, the project was officially shut down in early 2022. Source:
Fast Company
03-06-2025
- Business
- Fast Company
Is blind faith in AI a trap? These 5 steps use doubt as your best defense
In an April internal memo, Shopify CEO Tobi Lutke mandated 'using AI effectively' as a core expectation of employees. While AI boosts efficiency, leaders must use analytical skepticism to ensure that AI is safely and strategically integrated into their operations. Many organizations have faced expected challenges after implementation because AI, like any technology, can mislead. It may produce convincing but false predictions, such as 'data hallucinations' (fabricated facts or patterns) or biased outputs from flawed datasets. These risks can skew decision-making. To navigate this, leaders need analytical skepticism—a mindset of questioning AI results. They should ask critical questions: What's the source of this data? How reliable is the output? Where do human judgment and expertise still have value? This approach ensures AI aligns with business goals. Here are five actionable steps for leaders to use this trait to safely integrate AI into their business processes 1. START WITH THE PROBLEM, NOT THE TOOL Before diving into AI implementation, leaders must first ask: What problem are we solving? AI isn't a cure-all—it is a tool that demands a purpose. Without a crystal-clear grasp of the business challenge, organizations risk deploying AI that is misaligned, redundant, or worse, a shiny distraction. Only by pinning down the issue—be it inefficiencies, missed opportunities, or blind spots—can leaders judge if AI fits and how it slots into existing workflows. One of my clients implemented an AI platform to automate and optimize its ads to maximize its conversion rates for new customers. Using first-click attribution as a guide, we increased return on ad spend by 50% and reduced ad spend by 12%. This showed us that focusing on solving real problems drives meaningful business results. 2. SCRUTINIZE THE DATA Analytical skepticism demands more than good intentions. It requires robust data governance frameworks with systematic audits of sources, ongoing accuracy checks, and relevance assessments tailored to each AI use case. Organizations must thoroughly examine their data's origin, quality, and relevance before deploying AI solutions. The trustworthiness of AI outputs depends entirely on the integrity of its training data—a crucial fact often overlooked in the rush to innovate. As AI shapes high-stakes decisions—optimizing supply chains, forecasting market trends, or allocating resources—shoddy data quality can cascade into costly missteps. Consider Amazon's failed AI recruiting tool, which amplified discrimination after being trained on biased historical hiring data instead of improving fairness. Without rigorous vetting, even the most advanced systems can produce flawed insights, reinforcing problems rather than solving them. Skepticism means dissecting the data's story—its collection, context, and gaps—and demanding that it align with the task. Leaders must dig deeper than surface numbers. Is the data fresh enough? Does it reflect reality or distort it with hidden biases? For example, a retailer using pandemic-era sales data might overestimate demand, tying up capital in excess inventory. Only then can AI deliver decisions that hold up under real-world pressure. This isn't mere bookkeeping—it is a bulwark against 'garbage-in, garbage-out' disasters that undermine trust and tank ROI. AI can automate tasks and generate insights, but it is no substitute for human judgment. Oversight ensures outputs stay grounded and aligned with business goals, ensuring that AI amplifies expertise rather than overshadowing it. While AI quickly generates ad copy ideas, I recommend that my clients' copywriters and brand managers review all suggestions to catch off-brand creative. This human oversight ensures campaigns remain creative, consistent, and true to brand voice. Blending AI's speed with human insight keeps operations nimble and reliable, especially where complexity or stakes demand nuance over automation. This balance ensures AI delivers consistent performance while respecting the nuances of human judgment, particularly in areas that are too complex or sensitive for full automation. 4. FOSTER CONTINUOUS LEARNING AI implementation is not a one-time effort—it demands constant tuning. Leaders must build feedback loops to evaluate tools, question outputs, and refresh them with new data, ensuring AI stays sharp and reliable. By regularly questioning AI data, outputs, and tools, leaders ensure that their technology stack remains an effective and accurate resource. Zendesk nails this in customer service. Their Answer Bot's performance is tracked via agent reviews and customer ratings, driving upgrades that keep it on point. This relentless refinement, rooted in real-world input, ensures AI meets shifting demands without drifting into irrelevance. Skepticism fuels this cycle, turning scrutiny into progress. 5. ESTABLISH ETHICAL GUIDELINES AI technology is not neutral. If not managed carefully, it can entrench biases or spark ethical pitfalls. Leaders must set clear guidelines to ensure AI aligns with company values, focusing on fairness, transparency, and accountability. For example, our client's AI chatbot answered order inquiries instantly but didn't disclose it was AI, risking customer trust. The team made it say, 'I'm your AI assistant!' with an option to reach a human. This increased escalations by 5% but boosted satisfaction by 10%. Leaders should stay skeptical, checking AI outputs for bias or harm before scaling up. NAVIGATING AI WITH A HEALTHY DOSE OF SKEPTICISM AI is not a magic bullet—it is a tool that requires thoughtful, critical engagement to yield its full potential. Leaders who approach AI with analytical skepticism can successfully integrate this transformative technology into their business processes while mitigating risks like data errors and biased algorithms. By continuously questioning assumptions, verifying AI outputs, and ensuring that human judgment is incorporated into decision-making, leaders can navigate the AI landscape safely and strategically. The organizations that will thrive in the AI era are those that combine the power of technology with the wisdom of critical thinking.
National Post
02-06-2025
- Business
- National Post
Shopify scores win over Canada Revenue Agency in merchant-data case
Shopify Inc. has come out on top of a battle with the Canada Revenue Agency. Article content A federal court order issued Thursday shows Judge Guy Regimbald sided with the Canadian tech company, which was fighting the CRA's attempt to get more than six years of Shopify records. Article content The records were being sought in order to verify that Canadian merchants using Shopify software were obeying the Income Tax Act and the Excise Tax Act. Article content The CRA wanted the names of individuals who own Shopify accounts, their birthdates, addresses, phone numbers and their bank transit, institution and account numbers. Article content It also asked for their Shopify ID numbers, what type of store they ran, when their Shopify accounts were activated or closed and how many transactions and their value were made over the six-year period the CRA was interested in. Article content Some of the information had been requested by the Australian Tax Office, which wanted to ensure Shopify merchants were complying with the country's laws. A separate case Judge Regimbald presided over saw the CRA ask for court permission to obtain and send the records to Australia. Article content CRA spokesperson Sylvie Branch said the agency is aware of the court's decision and 'is currently analyzing the case details and associated information.' Article content Shopify pointed The Canadian Press to a post on X from its CEO, Tobi Lutke, who shared the outcome of his company's court battle and called the CRA's behaviour 'blatant overreach.' Article content Article content CRA demanded 6 years of Canadian merchant data from us. This felt like blatant overreach We took them to court and last Friday Justice Régimbald agreed with us. The court dismissed the request and called it '… unintelligible, incoherent, or otherwise beyond its understanding' — tobi lutke (@tobi) June 1, 2025 Article content Shopify fought the CRA in both cases when they were filed in 2023, insisting the group of merchants the agency wanted information for was 'overly broad and inconsistently defined.' Article content Article content The company also claimed a multilateral tax treaty being used to seek the information for Australia 'is without domestic force' when information about unnamed people is being requested. Article content Regimbald ultimately decided not to order Shopify to turn over the records to the CRA because he found the tax agency had not outlined an identifiable group of individuals whose data it wanted. Article content He said the court would not entertain a request to hand over information on unnamed parties 'that is unintelligible, incoherent, or otherwise beyond its understanding.' Article content
Fast Company
30-05-2025
- Business
- Fast Company
Shopify's user experience will soon ‘feel like sci-fi'
'Imagine an interface where you can quickly shift between talking, typing, clicking, and even drawing to instruct software, like moving around a whiteboard in a dynamic conversation,' Carl Rivera tells me. An experience in which users are not presented with a barrage of nested menus, but with a blank canvas that invites creativity aided by an artificial intelligence that knows everything there is to know about online and brick-and-mortar retail and marketing. A fluid interface that adapts and anticipates your needs, automating tasks and recommending actions like the most brilliant partner you could dream of. That's a dream in itself, but it isn't a fantasy; it's Rivera's future vision for Shopify. Rivera is the company's new Chief Design Officer and he believes that, in the very near future, the e-commerce platform's user experience is going to feel like sci-fi. Rivera joined Shopify through the 2018 acquisition of his startup TicTail. Right after that, he was key to launching Shop, the company's consumer-facing business. His new position directly responds to industry skepticism about design's relevance in an AI-driven landscape. In this time in which everyone is shifting to AI but almost nobody has a clear idea why, it makes sense that Shopify's founder Tobi Lutke thought he needed someone like Rivera helming that leading position. 'We're entering a new technological paradigm with AI,' Rivera says, emphasizing that now, more than ever, it is strategic for Shopify to have a clear design vision about how to implement artificial intelligence in a truly empowering way for every company, from small retail shops to corporate giants. The company wants to reimagine its user experience, transforming it into a powerful tool for designers and business people that is easier to use and saves more time than ever before. 'Half of the people are talking about design being dead because the programs can design for you,' he says. 'We take quite the opposite point of view at Shopify.' The final deadline for Fast Company's Brands That Matter Awards is this Friday, May 30, at 11:59 p.m. PT. Apply today.
