Hassan Nasser

How Engineering, Physics, and Rule Interpretation Collided in Formula 1 Formula 1 has always been a battleground not only for drivers, but for engineers who live in the gray areas of the rulebook. Some of the most iconic moments in F1 history were born not from outright dominance, but from clever interpretations of regulations. As Formula 1 prepares for the 2026 power unit regulations, a new technical controversy has emerged — one involving Mercedes and Red Bull Powertrains. At the center of it is a subtle but powerful engineering concept: thermal expansion and effective compression ratio. This article breaks down: 1. The 2026 F1 Power Unit Regulations — What Changed? For 2026, Formula 1 is introducing a radically updated hybrid power unit aimed at efficiency, sustainability, and cost control. One of the key changes affects the internal combustion engine (ICE): Why Compression Ratio Matters Compression ratio is the ratio between: Higher compression ratios generally mean: By reducing the allowed compression ratio from previous eras (≈18:1), the FIA aimed to cap power and efficiency gains. 2. The Key Detail: How the FIA Measures Compression Ratio Here’s where things become interesting. The FIA technical regulations specify how compression ratio is verified, and this verification is done: In other words, the compression ratio is checked when the engine is cold and static. The regulations do not currently mandate: This creates a small but critical gap between static legality and dynamic reality. 3. The Workaround: Designing for Thermal Expansion Mercedes and Red Bull are believed to have exploited this exact gap. The Core Idea All metals expand when heated — a fundamental law of physics known as thermal expansion. The equation governing this behavior is: ΔL = α · L₀ · ΔT Where: Inside an F1 engine, temperatures can exceed 500–700°C in the combustion chamber. What the Teams Likely Did Instead of designing an engine that stays geometrically identical across temperatures, engineers: But once the engine reaches operating temperature: All while remaining compliant during FIA inspection. 4. Effective vs Geometric Compression Ratio This distinction is crucial. Geometric Compression Ratio Effective Compression Ratio Mercedes and Red Bull appear to have optimized effective compression, not just geometric compression. This is similar in philosophy to flexible aerodynamic parts that pass static tests but behave differently at speed. 5. Why This Produces a Performance Advantage Even a small increase in compression ratio can yield: In F1 terms, this could mean: In an era where margins are measured in milliseconds, this is decisive. 6. Why Other Teams Are Concerned Manufacturers such as Ferrari, Audi, and Honda have reportedly raised concerns with the FIA, arguing that: However, under the current wording, the designs remain technically legal. 7. FIA’s Position — Legal, For Now The FIA has so far: Historically, this pattern is common in Formula 1: Examples include: 8. Why This Is Peak Formula 1 Engineering This workaround is not cheating — it is engineering excellence within constraints. It demonstrates: Formula 1 has never been about following rules blindly — it’s about understanding what the rules actually say. The Mercedes and Red Bull compression ratio workaround is a perfect example of how Formula 1 innovation evolves: Whether the FIA closes this loophole or not, one thing is certain: The fastest car is often built not just in the wind tunnel or on the dyno — but between the lines of the rulebook. 9. Why This Matters Beyond Formula 1 While this story unfolds at the highest level of motorsport, the underlying lessons extend far beyond Formula 1. In fact, the Mercedes and Red Bull workaround offers valuable insights for startups, R&D teams, and engineering-driven organizations in any industry. 1. Regulations Are Design Constraints — Not Design Killers In many industries — energy, automotive, medical devices, telecom, fintech — regulations are often viewed as blockers. Formula 1 shows the opposite: For startups, this means: 2. Test Conditions Matter as Much as Real-World Conditions One of the core lessons from this workaround is the difference between: Mercedes and Red Bull optimized performance where the engine actually operates, not just where it is inspected. For R&D teams: 3. Materials Science Is Often the Hidden Advantage This workaround was not about software tricks or exotic algorithms — it was about materials behaving differently under extreme conditions. Lessons for engineers: Many industries underinvest in materials R&D because the gains appear incremental — until they aren’t. 4. Optimize the System, Not the Specification A critical distinction highlighted by this case: Mercedes and Red Bull did not violate the compression ratio specification — they optimized the engine as a system. For startups and R&D teams: 5. Innovation Often Lives Between Disciplines This solution sits at the intersection of: Breakthroughs rarely come from one discipline alone. For engineering leaders: 6. The First Interpretation Advantage In Formula 1, the biggest gains often come early — before rules are clarified or closed. The same applies in business: This makes regulatory literacy a competitive asset, not an administrative task. Final Thoughts The Mercedes and Red Bull compression ratio workaround is more than an F1 story — it is a masterclass in applied engineering. It reminds us that: Whether you’re building race cars, hardware products, or deep-tech startups, the lesson is universal: True competitive advantage is often found not by breaking the rules — but by understanding them better than anyone else. Written for engineers, innovators, and R&D teams who believe progress happens where physics, creativity, and constraints collide.

3D Time-of-Flight (ToF) sensors have become increasingly essential in fields requiring depth sensing, spatial mapping, and object recognition. These sensors measure the distance between themselves and objects using the time it takes for a light pulse to travel to the object and return. With advancements in sensor technology, ToF sensors have gained prominence not only in consumer electronics but also in industrial applications and advanced fields like robotics, artificial intelligence (AI), and autonomous systems. In this blog, we will explore the technical workings of 3D ToF sensors, their significance in various industries, and their potential in future technologies such as AI and image recognition. How Do 3D Time-of-Flight Sensors Work?   3D Time-of-Flight (ToF) sensors measure the distance to a target object by calculating the time taken for a light signal, typically in the form of infrared (IR) or laser, to travel to the object and return to the sensor. This process is known as light pulse reflection or time-of-flight measurement. There are two primary methods used in ToF technology: direct time-of-flight and indirect time-of-flight. 1. Direct Time-of-Flight: In the direct ToF method, the sensor emits a modulated light signal (usually infrared) towards an object. The time it takes for the light to travel to the object and back is measured directly. This time is then converted into distance using the speed of light. This method is typically used in long-range ToF sensors due to its ability to cover large distances with high precision. 2. Indirect Time-of-Flight (Phase Shift Method): The indirect ToF method is more commonly used in 3D imaging and depth sensing. The sensor emits modulated light at a specific frequency and measures the phase shift between the emitted and received light. By comparing the phase shift, the sensor can determine the distance to the object with high accuracy, even at shorter ranges. This approach is widely used in applications where high-resolution depth maps and 3D imaging are required, such as in robotics and autonomous vehicles. Key Components of 3D ToF Sensors: Light Source: Infrared LEDs or laser diodes are used to emit modulated light. For 3D imaging, laser light is often preferred due to its ability to focus tightly on small objects. Detector: A photodiode or avalanche photodiode (APD) detects the reflected light. The detector measures the time it takes for the light to return to the sensor. Signal Processing Unit: This is the core of the sensor, responsible for calculating the time-of-flight and converting it into distance data. The signal processing unit processes the phase shift or time delay and outputs a 3D depth map or point cloud. Optics: Lenses or other optical components are used to focus the light on the target and direct the reflected light toward the detector. Importance of 3D ToF Sensors in Industrial Applications The industrial sector has adopted 3D ToF sensors in various applications due to their ability to provide accurate, real-time 3D data that enhances automation, monitoring, and decision-making. Some key industrial applications of ToF sensors include: 1. Robotics and Automation: In industrial robotics, ToF sensors are used for obstacle detection, path planning, and collision avoidance. By providing precise depth information in real-time, robots can autonomously navigate environments, identify objects, and interact with them in a safe and efficient manner. For example, in a warehouse, autonomous robots can use ToF sensors to avoid obstacles, stack products, and perform inventory management tasks. 2. Quality Control and Inspection: ToF sensors enable high-precision measurement and quality inspection in manufacturing. For example, in automotive production, these sensors are used to check the dimensions of parts, detect surface defects, and ensure proper alignment during assembly. The ability to measure at micrometer-level precision in real time ensures higher accuracy and reduces the need for manual inspection, thus improving productivity. 3. Machine Vision and Object Recognition: In industrial settings, machine vision systems leverage ToF sensors to detect objects, evaluate their size and shape, and ensure they meet quality standards. This is crucial for industries such as electronics manufacturing, where minute details and complex shapes must be inspected to maintain high quality. 4. Security and Surveillance: In industrial security, ToF sensors are used for intrusion detection and perimeter monitoring. By continuously scanning for movements and changes in the environment, they help detect unauthorized access or objects left behind in sensitive areas. 5. Augmented Reality (AR) and Virtual Reality (VR): In industries like architecture and design, ToF sensors are being integrated into AR and VR systems for spatial mapping and 3D modeling. By providing real-time 3D data, ToF sensors enable more immersive and interactive experiences, improving design processes and client presentations. The Future of 3D ToF Sensors in AI and Image Recognition The future of 3D ToF sensors lies in their integration with AI and machine learning to enhance image recognition, object detection, and interaction in both industrial and consumer applications. 1. AI-Powered Object Detection and Recognition: ToF sensors, when combined with AI algorithms, can provide advanced object recognition and gesture tracking. For example, in autonomous driving, AI algorithms can use depth data from ToF sensors to identify pedestrians, obstacles, and traffic signs, while AI models continuously improve the system’s ability to make decisions in complex environments. This integration makes ToF sensors indispensable in fields such as robotics, autonomous vehicles, and smart cities. 2. Human-Robot Interaction: ToF sensors are enabling natural human-robot interaction (HRI) by tracking gestures, poses, and movements. By integrating machine learning models, robots can understand and respond to human actions, facilitating tasks like assembly, medical assistance, and customer service. As AI and ToF sensors evolve, this interaction will become more intuitive and seamless. 3. 3D Imaging and Facial Recognition: In the realm of image recognition, ToF sensors can enhance facial recognition systems by providing accurate 3D face models. This helps in identifying individuals even in low-light or poor visibility conditions, making it suitable for security and authentication applications. 4. Enhancing Depth Data for AR/VR: In AR and VR, depth data from ToF sensors is crucial for creating realistic and immersive environments. By integrating AI,

Artificial intelligence is evolving at a rapid pace, but one of the biggest challenges remains: seamlessly integrating AI with external tools, apps, and data sources. Enter the Model Context Protocol (MCP), an open standard developed by Anthropic that aims to revolutionize the way AI interacts with external systems. What is MCP? MCP is a universal plug for AI, allowing it to connect effortlessly to various platforms—Google Drive, Slack, GitHub, and more—without requiring custom-built integrations. Instead of AI working in isolation, relying solely on pre-trained knowledge, MCP enables real-time access to relevant, dynamic data, significantly improving the AI’s effectiveness. Why is MCP a Game Changer? Traditional AI models often operate in silos, meaning they lack access to live, updated data from external sources. With MCP, AI can securely and efficiently retrieve information, making it more capable, up-to-date, and relevant. Here’s why MCP is a breakthrough: Smarter AI → Real-time access to fresh data leads to more accurate and context-aware responses. Faster Development → Developers no longer need to build custom API connections for each tool, saving time and resources. Open & Secure → As an open standard, MCP ensures a scalable, transparent, and secure approach to AI-tool interactions. How MCP is Already Making an Impact Big names in the tech world are already adopting MCP to enhance AI-powered applications. Companies like Replit, Sourcegraph, and Anthropic’s own Claude AI have integrated MCP to optimize their AI capabilities. For example: AI-Powered Code Assistants: AI can now fetch real-time GitHub data to provide smarter code suggestions. Enterprise Chatbots: AI-driven support systems can access Slack or Notion data to deliver instant, relevant responses. Enhanced AI Search: Instead of searching static documents, AI can pull real-time data from cloud storage solutions like Google Drive. What This Means for AI’s Future The introduction of MCP signals a paradigm shift in AI usability. No longer limited to pre-existing knowledge, AI can now dynamically interact with the tools we use daily, making it a more reliable and intelligent assistant across industries. For businesses, developers, and AI enthusiasts, this opens up a world of possibilities—from building smarter applications to automating complex workflows with ease. Final Thoughts MCP is not just a technical innovation; it’s a step toward a more connected AI ecosystem. With seamless, real-time data access, AI can finally move beyond static models and become a truly adaptive, intelligent assistant for individuals and businesses alike.

In the world of product development, structuring the design process is essential for creating successful, user-centered solutions. One of the most widely used frameworks for this is the Double Diamond, developed by the UK Design Council. This model provides a clear, structured approach to innovation, ensuring teams fully understand the problem before developing a solution. What is the Double Diamond? The Double Diamond is a four-phase framework that helps teams navigate the complexities of problem-solving and design thinking. It consists of two diamonds, representing divergent and convergent thinking. The First Diamond: Understanding the Problem Before jumping into solutions, it is crucial to define the problem correctly. This is where the first diamond comes into play. Discover (Divergent Thinking) – Research & Insights The goal is to explore the problem space widely without assumptions. Teams gather insights through user research, market analysis, and competitor benchmarking. Tools like surveys, interviews, and observational studies help uncover user pain points. Define (Convergent Thinking) – Problem Definition This phase refines the research findings to pinpoint the exact problem. Teams use frameworks such as personas, journey mapping, and problem statements to ensure clarity. The result is a well-defined challenge that will guide the ideation process. The Second Diamond: Developing Solutions Once the problem is clearly defined, teams can explore and refine possible solutions. Develop (Divergent Thinking) – Ideation & Prototyping Here, teams generate multiple solutions through brainstorming and creative thinking techniques. Rapid prototyping and early testing help explore different approaches. User feedback is gathered to validate and refine ideas. Deliver (Convergent Thinking) – Final Solution & Implementation The best solution is selected, refined, and prepared for launch. This phase includes final testing, deployment, and scaling of the product. Continuous monitoring ensures iterative improvements based on user feedback. Why Use the Double Diamond? The Double Diamond framework provides a structured yet flexible approach to product development. Its benefits include: User-Centered Design: Keeps users at the core of the process. Prevention of Premature Solutions: Ensures thorough problem understanding before solution development. Iterative Improvement: Encourages ongoing refinement through testing and feedback. Better Team Collaboration: Aligns different stakeholders towards a common goal.  

The ongoing discussion between OpenAI and DeepSeek has introduced a fascinating concept that will likely shape future conversations around AI development: distillation. OpenAI is accusing DeepSeek of distillation, a process where a smaller, more efficient language model is trained using responses from a larger, more advanced model. What Is Distillation? Distillation is a machine learning technique where knowledge from a large, complex model (teacher model) is transferred to a smaller, lightweight model (student model). This is done by training the student model on the output of the teacher model, rather than raw training data alone. The goal is to retain as much knowledge and capability as possible while reducing computational costs, memory usage, and latency. Key steps in distillation include: Generating Soft Labels: The larger model predicts probabilities for different possible outputs, providing richer supervision than traditional hard labels. Training the Smaller Model: The student model is trained on these soft labels, learning patterns in a way that approximates the teacher’s reasoning. Knowledge Transfer: The student model gradually approximates the teacher’s performance while being significantly more efficient and lightweight. This approach is particularly valuable in AI optimization because it allows for a balance between performance and efficiency, reducing redundancy while leveraging existing advancements. The Unique Optimization Path What makes this optimization approach interesting is its dual-model strategy. Instead of aiming for a single high-powered AI (like ChatGPT), DeepSeek is effectively creating two models: A fully-equipped, high-performance model akin to OpenAI’s GPT. A lightweight, cost-efficient model that delivers similar results with far fewer resources. This means that AI development isn’t just about making the most powerful model—it’s also about reducing complexity while maintaining performance. Projecting This Method in Research & Development How can we apply this principle in our own work, particularly in research and development? The idea of leveraging advanced insights to refine and streamline future iterations can be instrumental in optimizing innovation cycles. Here are some key ways this approach can shape R&D efforts: Experimentation & Prototyping: Instead of treating every prototype as a standalone iteration, we can introduce high-resource experimental models designed to extract detailed insights. These models could be more advanced, using additional computational power and sensors to collect in-depth data. Knowledge Transfer & Iteration: Once enough data is gathered from high-powered prototypes, we can distill that knowledge into lighter, more efficient versions of our systems, reducing costs without compromising on quality. AI & Automation in R&D: Applying a distillation-inspired workflow to machine learning and automation in research could accelerate discoveries by using AI-driven models to conduct extensive simulations before deploying optimized versions in real-world applications. Cross-Disciplinary Optimization: Whether in software, hardware, or engineering design, having an initial phase of data-heavy, resource-intensive research followed by an optimization phase can create more efficient and scalable solutions. By integrating this methodology, research teams can maximize efficiency while minimizing redundancy, creating innovative yet cost-effective products. The Practical Impact of Distillation in Product Development The applications of distillation go beyond AI and extend into real-world product development and industrial innovation. Some practical implementations include: Consumer Electronics: Companies can develop high-end flagship devices packed with cutting-edge technology, then use insights from user interactions to create more affordable versions without sacrificing key functionality. Autonomous Vehicles: Advanced sensor-heavy test vehicles can gather comprehensive data, which can then be used to optimize and streamline hardware in commercial vehicle models. Manufacturing & Supply Chain: Factories using advanced automation systems can analyze production workflows, enabling leaner, more cost-effective processes in smaller-scale operations. Retail & Market Analytics: High-data collection units can be deployed initially to gather detailed consumer insights, later leading to simpler, lower-cost tracking methods that still provide actionable data. By adapting distillation strategies across industries, organizations can achieve a balance between innovation and efficiency, ensuring that cutting-edge developments are not just theoretical but also practical and scalable. Final Thoughts While OpenAI’s concerns over distillation focus on competitive advantage and intellectual property, the underlying principle—using learned knowledge to optimize and streamline—presents a compelling approach to product development. As we continue to work on our own systems, we should explore ways to implement this dual-model strategy, leveraging high-performance insights to refine and optimize future iterations. Key Takeaway: Instead of always designing for maximum power, consider a two-tiered approach: develop a high-powered learning system first, then use its insights to create a cost-effective, efficient model that delivers comparable results. References: OpenAI’s claims against DeepSeek: Financial Times Understanding Knowledge Distillation: arXiv: Distilling the Knowledge in a Neural Network Applications of Model Compression in AI: arXiv: Quantifying the Knowledge in a DNN to Explain Knowledge Distillation for Classification AI Optimization Strategies: Neptune.ai: Knowledge Distillation