Greg Coquillo

High-performance AI is not built in a vacuum but on strong foundational engineering. I spent the holiday break completing my 7th Databricks certification: Machine Learning Associate. While it is tempting to stay at a strategic 30k foot view, I believe an AI executive must stay grounded in the technical reality their teams face every day. In large-scale corporate environments, architecture is where the time dimension becomes critical. A shortcut taken today might deliver a quick win, but the right architectural choices ensure that you are building a scalable asset for the next five years rather than technical debt that paralyzes the organization in eighteen months. Why I believe technical rigor is a business imperative: 🔹 Precision in Execution: I was reminded this week how a simple shift from classic Python UDFs to Pandas UDFs on Spark can transform performance. It sounds like a detail, but it can be the difference between a sluggish pipeline and a cost-optimized engine that potentially saves 100k Euro in annual cloud spend. 🔹 Trust and Reliability: In a recent production use case, we achieved >92% prediction accuracy and cut processing time from hours to seconds. You do not reach that scale by accident. You reach it by validating production workflows against the latest platform capabilities. 🔹Empowering the Team: As a leader, staying current is not about doing the work for your team. It is about setting the standard. Practical experience with frameworks like ISO 27001 and 42001 provides the safety rails for high-speed delivery instead of bureaucratic roadblocks. I have found that the most effective technical leadership ensures engineering precision translates eventually into documented business impact. The objective is rarely just to build a model but to architect the industrial-grade platforms that makes those solutions sustainable and reliable for the long term. I am curious: What foundations are you laying today to ensure your Data & AI strategy scales effectively over the next years?

78

16 Comments

Apple Intelligence and AI Foundational Model: The Great On-Device AI Trade-Off Apple Intelligence is a fundamental architectural commitment: bringing foundational model power to the edge. This philosophical pivot changes the calculus of privacy, performance, and platform lock-in. Executives and AI professionals must grasp the critical trade-offs of local-first processing. The Promise: Why On-Device is the Future Running a dedicated foundational model locally (A17 Pro/M1+) is powerful, touching on three core pillars: Data Privacy (Zero Exposure): Data never leaves the device or hits verifiable Private Cloud Compute (PCC). PCC servers run on custom, verified silicon, ensuring complete confidentiality. Low Latency (Instantaneous): Local processing eliminates network delays, offering superior UX for real-time tasks like summarization. Performance is limited only by the Neural Engine’s throughput. Deep Personalization (Contextual): Immediate access to personal data (messages, photos) enables context-rich assistance cloud models can’t match. It utilizes specific app APIs for retrieval. The Reality Check: The Architectural Limitations While privacy and speed are clear benefits, on-device AI introduces significant constraints: Capability Ceiling (Scale Limitation): Models are constrained by device memory and thermals, limiting complex reasoning. Enterprise-grade or highly creative tasks still require large cloud LLMs (GPT-4o, etc.). Hardware Dependence (Forced Refresh): Only devices with A17 Pro / M1+ Neural Engines are supported. This forces accelerated device refresh cycles for companies, increasing CapEx. Hybrid Bottleneck (PCC Reliance): Many key features rely on Private Cloud Compute (PCC). If the network stalls, complex requests fail, making the "off-the-grid" promise misleading for mission-critical tasks. My Takeaway: Apple is not winning the parameter race; it's winning the trust race. By betting on a 'privacy-first, local-first' architecture, Apple forces the industry to justify data centralization. The true test lies not in speed, but in how quickly developers adopt the platform and how on-device constraints influence application complexity. This establishes a powerful tension between capability and confidentiality. We are entering a new era where the "best" AI is defined less by its raw intellect and more by its proximity to the user. I explored these on-device capabilities by creating a small PoC to generate code and images locally(Apple Silicon MacBook Pro) with the Apple Foundational AI model( MLX-based LLM with a Stable Diffusion XL pipeline). https://lnkd.in/e_y-VhzS What specific enterprise application do you foresee being radically transformed by on-device, privacy-preserving AI in the next 12 months? Drop a comment and share your view. #AppleIntelligence #OnDeviceAI #FoundationalModels #AIStrategy #TechLeadership #MachineLearning #GenerativeAI #AIEthics

20

2 Comments

AMD and Oracle Cloud (OCI) to accelerate drug discovery with Absci, a clinical-stage biotech firm advancing novel therapeutics design with Gen AI. Learn more: https://lnkd.in/g5_Ytjj4 #TogetherWeAdvance_life

42

1 Comment

I enjoyed speaking in the #GoogleMITWorkSummit panel discussion on "Beyond boosting productivity: How AI adopters are leveraging AI technology to achieve strategic goals". Thanks to the moderator, Lisa Gevelber. The most critical takeaway is that unlocking the true economic value of generative AI is not about substituting old processes; it’s about reimagining the entire factory floor. Leaders must move beyond simple productivity gains to re-envisioning our core business processes—together with people doing the work—to successfully reap the benefits of this technology. This is a generational opportunity that demands strategic adaptation. #GoogleMITWorkSummit

106

4 Comments

Gemini 3 is here! It was incredible reading about Google's new model. It's feels far more than a simple performance bump—it signifies a real shift where AI moves into being a genuine “thought partner”. Few features that stood out: • State-of-the-Art Reasoning: I'm excited that this model is built to genuinely grasp depth and nuance, which means it's phenomenal at "reading the room" for complex problems and is setting a new standard for intelligence. • Advanced Agentic Power: What really caught my attention is Google's claim that this is their most powerful agentic and vibe-coding model yet. It enables planning and executing complex, multi-step workflows, which is a massive leap forward for productivity. • Broad Multimodality: I was amazed to read it's claimed as the best model in the world for multimodal understanding. It can seamlessly synthesize information—vision, code, and text—helping us learn, build, and plan complex tasks in totally new ways. Looking at these features, we are entering a new era of intelligence. I absolutely cannot wait to see what everyone builds with this! #Gemini3 #GoogleAI #AIInnovation #FutureOfWork #LLM #AI https://lnkd.in/guMDTxAU

114

4 Comments

The AI Oracle Has Spoken: Andrew Ng’s Bold Predictions for the Future! I just dove into this fascinating deep dive on Substack—check it out https://lnkd.in/eNY5GDbf 1/ Agentic AI is the New King Forget scaling massive models that cost billions. The agentic AI market is set to explode from $5.1B to $69B by 2032—a 13x growth in just 7 years! Ng’s revolutionary proof? Smaller models with agentic workflows (think self-critiquing AI, API integration, task planning, and multi-agent collaboration) outperform giants. Even JPMorgan has slashed costs by 30% using this approach. 2/ Military AI: The Next Gold Rush In February 2025, Ng stunned Silicon Valley with, “I’m glad Google changed its stance on AI weapons.” His portfolio companies are already saving lives with autonomous drones—proof that defense is a frontier ripe for innovation. 3/ AGI? Decades Away Ng’s simple test for agi: “Until companies fire ALL intellectual workers, AGI hasn’t arrived.” For now, his focus is on solving “boring, profitable problems”—a pragmatic approach driving real impact. 4/ China’s Open-Source Edge Ng warns, “There’s now a path for China to surpass the U.S. in AI”—not with bigger models, but with speed. Smart founders are leveraging Chinese open-source models at 1/10th the cost, shaking up the global race. 5/ Small Specialized Models Are Winning The SLM market is projected to grow from $930M to $5.45B by 2032, fueled by a 90% drop in token prices and the edge computing boom ($378B by 2028). Imagine medical AI on your phone, factory AI on $99 devices, or retail AI with zero latency—no cloud costs, privacy concerns, or internet needed. The infrastructure for the next wave of AI millionaires is here. But here’s the catch: TRUST is the missing piece. Customers demand trustworthy AI decisions, regulators require explainable models, and stakeholders need transparency. Most deployments lack the governance frameworks to deliver this. The winners? Those deploying validated models with robust monitoring—separating leaders from followers. #Innovation #EdgeComputing #Leadership

10

2 Comments