Kayla Leung

Here's an email from October that I never thought I'd write: asking a prof for an extension to help with juggling school, a project in Boston, a project in SF, and interviews--all in one week. I spent four days at Harvard for a hackathon, landed back at college on Monday to tackle school work, had two interviews on Wednesday and Thursday, and left Thursday for a hackathon at Berkeley. It was overwhelming, and I realized that you really can't do it all. You have to prioritize (and thus also deprioritize). Here’s how I approached it: 1. Interviews (I need a job lol) 2. Hackathons (I wanna win 🤩) 3. College (apparently, I need a degree, lol) Having too much to do--and being unable to do it all perfectly--is honestly what I find exciting about life. It keeps you on your toes. And it forces you to continually learn and grow.

403

39 Comments

Highly recommended book...

9

We’re hiring at Palantir! Apply now: https://lnkd.in/gbvYh4fr It’s difficult to overestimate how transformative the first few months at Palantir are for a new Forward-Deployed Engineer. As an FDE, you’re responsible for it all—from going onsite and scoping out your customer’s toughest challenges to building out the workflows that will drive them to overcome these challenges. Because of Palantir’s flat and mission-driven culture, these responsibilities really do fall on you. The growth I've experienced at Palantir from wearing so many hats and having radical ownership over my team's workflows is both unique and unparalleled. I’ve had the privilege of working with the sharpest people and most fascinating products, all while solving highly complex and impactful issues. What does it take to become an operating system for the modern enterprise? How much impact can you have on some of the world’s most important institutions, all while solving their most interesting problems? How do you zero-to-one solutions that tackle a business’s or government’s most mission-critical workflows? If any of this is of interest to you, let’s chat about becoming a Palantir FDE.

36

1 Comment

My time as a research engineering intern at Google DeepMind has come to a close! As someone whose world revolves around networked, computer, and distributed systems, I never imagined I'd land on my dream team – the model partitioning team! This summer, I worked on PartIR's automatic model partitioning capabilities(https://lnkd.in/egp5Ygvy), diving deeper into distributed computing and exploring applications in reinforcement learning. It was definitely a challenging project, but I picked up so many invaluable skills, from software engineering to research in an industry setting. I’d like to give a huge thank you to Sami Alabed, Masha Samsikova, and the entire model partitioning team for making me feel so welcome and helping me successfully complete my project! I’m also thrilled that my final presentation was well received (and even got some laughs!), helping many at GDM understand a complex topic. Outside of my project, I had the chance to meet so many fascinating people across different teams and, of course, the other interns! A big thanks to the GDM Academy organizers, Charlotte Colson and Claudia Pope, for their incredible support in running such a fantastic program while I was balancing my thesis. And to Lucia Lopez Rivilla and Ioana Mihăilescu, former GDM Academy interns, for helping me find my way around the company! This summer wasn’t without its challenges, though. I finally addressed the anxiety issues I’d been struggling with since my teenage years, as they were impacting my studies and relationships. Unfortunately, I wasn’t able to attend my final year exams as I had hoped. However, I'm so grateful to UCL's Computer Science team and Faculty of Engineering for giving me the flexibility to extend my studies by a year, without penalty, and extending my scholarship. I couldn’t be more thankful for their support! All in all, this experience has solidified my determination to pursue research. In my third year, I discovered my passion for networked systems, and during my time at AMD, I knew I wanted to work on distributed systems. Now, after my time at GDM, I’m confident that distributed and networked systems are my future – and with the added experience of AI, I feel more prepared to navigate the machine learning-filled computer science landscape. Here’s to a (hopefully) restful but equally exciting year ahead! See you soon, Morgane 🦆

219

16 Comments

Meta released their New Grad software engineer and 2025 summer internship job postings! Here they are: New Grad - https://lnkd.in/e6KcE-wX Summer 2025 Internship - https://lnkd.in/e_9yMFyR Meta University Internship (for Freshman / Sophomores) - https://lnkd.in/edA9Ewwy I’ve been loving my time at Meta and highly recommend it. Here you work with, learn from, and are mentored by the best engineers in the industry. Hurry and find someone who can refer you and then get your application in ASAP! Heads up, feel free to reach out to me to ask for a referral, but with the large number of requests and limited referrals I can give, I can’t promise anything. But I’ll do my best to help in other ways! One of the ways I can help is by sharing spreadsheets I am maintaining that are full of internship and new grad opportunities at companies that would be at the top of my list if I was recruiting right now. The spreadsheets are in the same template that I used to track my application progress when I was recruiting. They also include some helpful tips on how to secure referrals. Send a connection request and comment “interested” on this post if you want me to dm it over to you. Good luck out there. #SoftwareEngineer #coding #recruiting

143

497 Comments

Ever had a friend who can’t stop tweaking their fantasy football team, hoping to get the perfect lineup? That’s what tuning a machine learning model can feel like. Let’s talk about hyperparameter tuning Day 17: Machine Learning Interview Questions What is hyperparameter tuning, and why is it essential for improving model performance? Answer: Hyperparameter tuning is the process of finding the best set of hyperparameters (settings) for a machine learning model to maximize its performance. Unlike model parameters, hyperparameters are values set before training, such as the learning rate, the number of trees in a random forest, or the regularization strength in logistic regression. Here’s why it matters: 1. Finding the Optimal Settings: Models can behave very differently depending on the hyperparameters. Tuning helps you find the sweet spot for your model’s performance, ensuring it generalizes well to new data. 2. Avoiding Underfitting and Overfitting: Hyperparameter tuning helps you balance the complexity of your model. Too simple, and the model underfits the data. Too complex, and it overfits. Tuning helps you strike the right balance. Methods for Hyperparameter Tuning: Grid Search: A brute-force approach where you try all possible combinations of hyperparameters and select the one that gives the best performance. Random Search: Instead of trying all combinations, you randomly sample from the hyperparameter space. This is faster and often just as effective as grid search. Bayesian Optimization: A smarter way of searching for the best hyperparameters by learning from past results, allowing you to focus on promising regions of the hyperparameter space.

2

Excited to share a preview of the AI Prescreening Assistant I’ve been developing! This tool prescreens candidates via calls and has incredible potential in Customer Support, Sales, and Marketing. Current Architecture: Twilio for calls Speech-to-Text models for speech to text conversion. Text-to-Speech models for text to speech conversion. LLM's for response generation. With OpenAI Releasing Real-time API, this entire process will shift to a seamless Speech-to-Speech model, revolutionizing real-time conversations! DM if you want be beta Tester of Prescreening Assistant. :) #AI #Twilio #VoiceAI #OpenAI #Innovation #Prescreening

61

9 Comments

📣 Happy to announce that I have accepted my New Grad Software Engineering offer at Bloomberg, NYC. Coming from Tokyo as an international student, this is DREAM COME TRUE, being able to work in the US in this tough tech economy. ⌛ THIS HAS BEEN A LONG JOURNEY — thinking back where I have dream about being a pastry chef 👨‍🍳 and open my own cake shop to end up being a software engineer in the US. 🌠 Just a little bit of my journey.  I started English in elementary school, going to cram school, dreaming about college in the US. I used to think life feels perfect but something was missing - something that would push me to grow. My instinctive thought made me wanna go to study abroad so I took one big step to go to boarding school in Canada. Biggest challenge in my life was to just fly to a new country alone, and excel there. I was just struggling to even knock someone’s door and become close friends or talk in the circle of all the native speakers. I almost stopped having hope and confidence for myself, but at one point, I realized the core of my work ethic and my personality hasn’t changed much. I just kept working hard, and got amazing friends and good grades, and got into my dream school at UC Berkeley. My next challenge was to pursue CS, coming here as an intended economics/business major, surrounded by intelligent genius people. I felt like I was so behind of everyone when even nothing started yet, and it hit so hard when I applied to CS clubs and rejected. But this time too, I didn’t give up. I self studied with Youtube and online classes. In Spring 2022, I applied 15 clubs with so many rounds of interviews, and got into 4 of them. I hit to the biggest wall in college: getting summer internship. I submitted 600 applications and I almost failed everything … I was not good enough for coding and interviews. But I didn’t give up this time too. I just kept applying, trying, practicing, preparing repeatedly, and finally got two offers at the end. ✔ Now finishing my college degree with 5 internships, 3 contract client projects, and 3 freelance job experience, which I never thought I would have done. "Work hard and never give up" — this sounds so cliche and vague. Yet, having confidence and hope is the only way to go through depressing time in life. I don’t know how many times I felt like I’m stuck in the dark never-ending room. But life is also a series of dot. Sometimes, that struggle, cry, depression, hard work may just sound unfortunate event for you, but I believe that every single dot that happened in my life, no matter good or bad, have led me who I am and where I am, so I def feel stronger going forward pursuing something different. And I hope this message pushes someone's back too. Thank you everyone who have been in this journey, whether they are my teachers, mentors, friends in Japan, Canada and US, and my parents. I would not stop working hard — hopefully keep going but this time with more confidence. 

2,246

157 Comments

This Quick Summer Session II was full of up and downs and unexpected turns, but with the help of my teammates Frankie Chiu, Pratham Joshi, and last but not least Isaac Lin, We rode the lighting effortlessly. Proud to Present: Our Robocar Project - MAE148 Team 1 Final Project! We applied advanced robotics concepts such as SLAM, AMCL localization, and obstacle detection using the OAK-D-LITE and LIDAR sensors. Our project integrates cutting-edge technologies across both ROS1 and ROS2 systems to achieve seamless communication between the car's hardware components and software algorithms. We tackled real-world challenges like odometry drift, precise path-planning, and efficient 3D object detection. Key features of our project include: ~Simultaneous Localization and Mapping (SLAM) for real-time map generation ~ DepthAI for 3D/2D object detection integrated with ROS1 ~Path-planning algorithms for autonomous navigation ~Cross-communication between ROS1 and ROS2 environments This project has been a valuable hands-on experience, blending engineering skills across electrical, mechanical, and computer systems. The practical knowledge I've gained through this collaboration will be instrumental as I move forward in my career. Looking forward to applying these skills in future opportunities! #Robotics #ROS #AutonomousSystems #Localization #PathPlanning #Engineering #Teamwork https://lnkd.in/daqhZejk Special thanks to Frankie Chiu for giving the GitHub page a structured look

10

1 Comment

🚀 Introducing SpectrumX: Next-Generation Multi-Device Data Acquisition Software 🚀 After 8 months of development, I’m excited to share SpectrumX (Alpha), a cross-platform, high-performance, and fully reactive solution built from the ground up with TypeScript, React, and Rust. Key features: ⚙️ Device-API: Developers can write their own device plugins in TypeScript and seamlessly integrate them into the Spectrum runtime. 📊 WebGL charting: Capable of rendering over 10 million points per second on average hardware, maintaining smooth performance at 40 FPS. 🧵 Multi-threaded architecture: Handles data acquisition, database operations, and real-time calculations with optimal performance and stability. 🖥️ Cross-platform support: Runs on Windows, Mac, and Linux, ensuring wide accessibility for different environments. 🎨 Custom accessible UI components: Built with React for a responsive and user-friendly experience. With SpectrumX, we are trying to push the boundaries of what real-time data acquisition and analysis software can achieve. Check out the promo video to see it in action! #softwaredevelopment #typescript #rust #reactjs #webgl #dataacquisition #crossplatform

17

1 Comment

Doordash just published one of the best blog posts I've read in a long while on how they use AI and LLMs in software engineering! 🚀 At this point, we all are well aware of AI advancements, and LLMs like ChatGPT, LLama, etc. While they have been immensely useful to me till now in writing emails, and writing unit tests, this blog post by Doordash highlighted how they use LLMs in their systems to better classify products, to identify duplicates and improve user experience. For example, one of the problems highlighted in the blog post is how do you identify two different product names are the same product, for example, “Corona Extra Mexican Lager (12 oz x 12 ct)” sold by Safeway is the same product as “Corona Extra Mexican Lager Beer Bottles, 12 pk, 12 fl oz” sold by BevMo. They leveraged a mix of software systems and AI models to identify and understand their products. https://lnkd.in/gtjM4NgB Kudos to Steven Xu and Chaitanya (Sree) Vadrevu for writing such an interesting post! 💬 Want to delve deeper into such intriguing tech discussions? Follow me on LinkedIn for more insightful updates, or check out my Medium blogs - https://lnkd.in/gBmNjZTG for in-depth technical dives! Let's keep the conversation going! 🔍✨

58

2 Comments

Last weekend I had the opportunity to go to UC Berkeley's AI Hackathon. I still can't believe that in just 27 hours, we hacked together a customer service factory (more on that later). Shoutout to Richard Ky, Ejiro Igun, and Ranveer Singh for being such amazing team members! I had a great time working with you all and learned so much! 𝐖𝐡𝐚𝐭 𝐰𝐞 𝐝𝐢𝐝 We created an application that creates customer service agents to solve the shortage of good customer service representatives. Just enter your website's URL, and our application gets to work generating a customer service agent. Once ready, our application generates a phone number that customers can use to talk to the new agent. Our LLM-based agent will handle all incoming calls and answer in a friendly, human way based on the information available on your website. I put together the speech-to-text and text-to-speech pipeline using Deepgram and LMNT, added phone support by using Twilio, and worked hard to get our scraping and RAG pipelines working efficiently. Groq's ultra-low latency LLMs really helped our application feel like a conversable human. I still can't believe we did that in just 27 hours! You can check out our demo, where we create a customer service agent for the hackathon and call in with questions, below!

16

Thank you, Robert and Justin, for insightful discussion. In the 𝐌𝐞𝐚𝐧𝐂𝐚𝐜𝐡𝐞 study, we argue that traditional cache hit and miss rates are misleading for 𝒔𝒆𝒎𝒂𝒏𝒕𝒊𝒄 𝒄𝒂𝒄𝒉𝒊𝒏𝒈 and should not be used. MeanCache is a user-centric semantic cache for LLM-based web services (e.g., ChatGPT). It uses Federated Learning to fine-tune smaller embedding models, enhancing query semantic matching and handling context queries effectively. 𝐏𝐚𝐩𝐞𝐫: https://lnkd.in/e9PB6qie

13

Diving into AI: My Exciting Journey with Headstarter AI Fellowship! 🚀 Just wrapped up my first week of virtual meetings with the Headstarter AI team, and I'm buzzing with excitement! Though the fellowship officially kicks off on July 22nd, I've been soaking up knowledge daily from co-founders Yasin Ehsan 🚀 and Faizan Ahmed. Here's what we've covered so far: • Qualities of great software engineers • Networking tips for tech professionals • Standing out in the competitive SWE field I'm gearing up to build 5 AI projects over the next 7 weeks, culminating in a final project that I can't wait to share. Looking forward to collaborating with my team and attending some in-person meetups soon! What's your most valuable takeaway from a recent learning experience? Share in the comments! #HeadstarterAI #SoftwareEngineering #AIProjects #TechLearning 👨‍💻🌟

8

1 Comment

💡 Working Smarter is a new podcast from Dropbox about how AI is changing the way we work and get stuff done—not in the future, but today. 🤖 On our first episode, hear robotics expert Kate Darling talk about how to think about our relationship with our machines. Click to listen below!

1

Last week at HackMIT, my team (Sanskriti Akhoury, Krish Modi, and Aly Shariff) and I built Runk: an AI-powered e-commerce platform that understands your preferences to display relevant reviews and products. How do we know your taste? Using autoencoders, we assign each user a profile embedding, which our own model updates as you leave more reviews. With these profile embeddings, we rearrange e-commerce search results to display products with the most positive reviews left by people most similar to you first. We also built a Google Chrome extension to interface Runk onto pre-existing sites like Amazon to re-rank these reviews based profile similarity. A huge thank you to HackMIT for making this hackathon an amazing experience. I also enjoyed meeting new people at Massachusetts Institute of Technology, Harvard University and Northeastern University! 👾https://lnkd.in/eyhU3nvK Outside of this hackathon, here’s what I’ve been up to this summer: • Built Guardian at HackThe6ix - an AI wearable for SOS situations. With a Raspberry Pi, Gemini Flash, and Hume AI, it contextually analyzes audio and video of emergencies and sends a summary to first responders. Collaborators: Sanskriti Akhoury, Daniel Hong, and Brian Zhang. 👾https://lnkd.in/ei5-Gu46 • Built Siren - an AI-powered dashcam. It analyzes red light camera runs, illegal turns, speeding, and other driving behaviors using YOLOv8 and algorithms built with reverse geocoding. 👾https://lnkd.in/eKMs2x5t • Joined Mimrr as a Machine Learning Engineer to work on their LLMs for automating code documentation. • Joined WATonomous as an Autonomous Software Engineer to work on self-driving cars. You can view these projects on my new website! 👾https://lnkd.in/e-hWsAxB

987

20 Comments