About
Activity
2K followers
-
Zhong Zheng liked thisZhong Zheng liked thisThree months ago I joined Pacific Blue as Head of AI and Data. Time to say why. Jensen Huang describes AI as a five-layer cake: energy at the bottom, then chips, infrastructure, models, and applications at the top. Energy is the layer with nothing beneath it. Every token generated anywhere in the world traces back to electrons. He calls it the binding constraint on intelligence itself. I find the position of Energy genuinely fascinating. It sits at the foundation of AI, and it's also one of the industries with the most to gain from it. Forecasting, asset performance, market analysis, the daily grind of getting things done. Both ends of the cake at once. Not many industries can say that. I spent most of my career in tech, where AI has already rewritten how everything works. I believe the next decade belongs to traditional industries — the ones that generate power, move goods, build things — where AI meets physical reality and real value gets created. And it's work that gets built with the people who understand these systems better than any model does. What convinced me to join was Pacific Blue's 100% renewable vision, and what has kept me energised since is the people. Deep operators, generous with their knowledge, genuinely curious about what's possible. The best transformations start with talent like that, not with technology. We're at the cutting edge of something big. Excited for what we build next. #AI #Energy #RenewableEnergy #DataAndAI #PacificBlue
-
Zhong Zheng liked thisZhong Zheng liked thisIf you're interested in improving the quality of AI enabled products in both Square and CashApp, we're hiring! https://lnkd.in/gVy-2tsfCareers - Senior Machine Learning Engineer, Applied AI QualityCareers - Senior Machine Learning Engineer, Applied AI Quality
-
Zhong Zheng liked thisZhong Zheng liked thisWe chased autonomy for a year. Then the bill showed up. When I built out our AI-native team, the mandate was simple: make feature development as autonomous as humanly possible. Stop babysitting the agent. Let it plan, build, gate, and ship. So we pushed the boundaries. Skills. Plugins. Hooks. Workflows. We turned spec-driven development into something that runs with less and less human steering every week. It worked — features that took days now move in hours. But every advantage breeds a new disadvantage. The more autonomous the agent, the less visible the cost. Every run spends real money in tokens, quietly, across every repo and every engineer. We'd optimized for can it ship — and stopped asking what did it cost, and what was it worth? That's the next frontier of engineering optimization. Not just velocity. Token economics, baked into the decision-making itself. So today I would like to interduce 100xPrism - rebranding 100x-dev , with token + value monitoring built in: 🔹 100x-tokens — one machine-wide dashboard: input/output/cache split, startup-bloat meter, $ cost by project, model, and day 🔹 100x-value — because tokens measure cost; this measures what actually shipped for that spend A prism takes one beam and reveals everything inside it. That's the idea — take one stream of agent work and make its true cost and value visible to the whole team. Because watching the spend isn't one person's job. It's everybody's responsibility. Autonomy got us here. Economics is how we go further. Would love to hear what's your team doing to keep AI dev cost-aware? #AIEngineering #DeveloperTools #AgenticAI #ClaudeCode #EngineeringLeadership
-
Zhong Zheng liked thisI'm really excited about Buzz, a tool my team has been working on. The way I think about it: most of our agentic harnesses are one human to one-or-more agents. Buzz is many-to-many, putting collaboration front and center. It's kind of brain bending... but in a good way!Zhong Zheng liked thisBeen working on something fun at Block called Buzz and this is the first, small way we’re sharing it. A workspace built for teams of humans and agents. https://lnkd.in/dXMASuKw Apache 2, Rust, Nostr, ACP, MCP
-
Zhong Zheng liked thisI left Block last May after 9 years to take a year off — to relax, travel, and figure out what I wanted to work on next. After a lot of reflection, the clear answer is AI risk reduction: I'm deeply concerned about how AI advancements could outpace our ability to manage their risks. AI-developed bioweapons are one class of AI risk that's worth taking seriously. That's why I'm joining Valthos to build AI systems that detect, characterize, and respond to biological threats faster than they can spread. I'm thrilled to be working on an urgent and important problem alongside an exceptional team. If this sounds like a problem you want to work on, we're hiring — reach out!Zhong Zheng liked thisWe’re thrilled to welcome Mackenzie Shelley to Valthos as a Member of Technical Staff on our engineering team. At Valthos, Mackenzie will help build the platform that compresses the reaction time of biodefense. Mackenzie joins us from Block (formerly Square), where she spent nearly a decade building the data and AI infrastructure that powers analytics and machine learning across the company. Most recently, she was a tech lead for the data and AI platform organization, where she drove foundational work spanning Block’s ML platform, data platforms, and streaming systems. Earlier at Square, she worked on the Risk team, building the machine learning systems that detect and prevent fraud on the platform. She holds a B.S.E. in Computer Science from the University of Michigan.
-
Zhong Zheng liked thisZhong Zheng liked thisI've been building data platforms since Yahoo ran HDFS on bare metal. 20 years. Petabytes of data. Yahoo → VMware → LendingClub → Experian → Udemy. Every budget cycle, we defended our existence instead of proving our impact. Because we always started with the wrong question: "How do we store and move data efficiently?" The right question is: "What revenue moat does this data create — and can an agent act on it today?" In 2026, your primary consumer is no longer a human analyst. It's an agent. And that changes everything about how you design the platform. 👇
-
Zhong Zheng liked thisZhong Zheng liked thisGoose has a new nest! https://lnkd.in/gYC2bVJq Last fall, we announced that Goose, the open source agentic harness started by Block, is now under the purview of the Linux Foundation's Agentic AI Foundation, alongside MCP and Agents.md. As of this week, the code has officially moved! This move reinforces our vision of Goose as not only an open source agentic harness but one which can become the reference implementation for standards and a foundation for experimentation and discovery. Thank you to all of the Goose contributors and the folks at the AAIF and Linux Foundation who helped make this happen!GitHub - aaif-goose/goose: an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLMGitHub - aaif-goose/goose: an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
-
Zhong Zheng liked thisZhong Zheng liked thisSpent a few late nights deep in AI workflows — generating code, refactoring, reviewing endless .md files. At some point, a realization hit me. The tools are impressive. The speed is addictive. The output feels almost magical. But the tools aren’t the advantage. Models will get better. Frameworks will change. Today’s “must-have” stack will be replaced. What doesn’t get replaced? The data you own. The feedback loops you build. The insights your system learns over time. That’s the real leverage. AI accelerates execution. Data compounds value. And compounding wins.
Experience & Education
Volunteer Experience
-
Volunteer
Second Harvest Food Bank of Santa Clara and San Mateo Counties
- 2 months
Social Services
Helped to pick up and pack the pears.
-
Volunteer
Habitat for Humanity Greater San Francisco
- 2 months
Social Services
Helped to paint a house
Courses
-
CSCI561 Artificial Intelligence
-
-
CSCI570 Algorithm
-
-
CSCI571 Web Technology
-
-
CSCI578 Software Architechture
-
-
CSCI580 3D Modeling
-
-
CSCI585 Database
-
-
CSCI588 User Interface
-
Projects
-
Email subscription system and Cron crawler
See projectTech Stack: GAE, python, bootstrap
-- Click the link, and find something interesting.
-- https://zhong-1255.appspot.com/ -
Scientific Search Engine
Tech Stack: java, mangodb
-- Crawled the scientific data from several scientific website and parse these data
-- Developed exact de-duplicate plugin and near de-duplicate plugin for Nutch
-- Indexed all the data and implemented PageRank algorithm
-- Implemented content based search and analysed the search result to answer some scientific questions -
Campus Spatial Search Software
Tech Stack: java, mysql
-- Developed java application with spatial database
-- Stored geometry information of USC campus in oracle spatial database, including locations of buildings and students
-- Did queries for students and buildings locations information
-
2D Side-Scrolling Game on iPhone
-
Tech Stack: cocos2d, objective-c
-- Made a popular side-scrolling game in which the hero can run, jump, fly, rush and fall to get as far as possible, based on Cocos2DOther creatorsSee project -
Motorcycle Market
-
Tech Stack: php, bootstrap, mysql, javascript, html, css
-- Developed an E-Commerce Website
-- Implemented user information system, cart and order management system and product recommendation system
-- Implemented employer-manager information management system, and customer history query function
-- Implemented the project by CodeIgniter
Other creatorsSee project
Languages
-
English
Professional working proficiency
-
Chinese
Native or bilingual proficiency
Recommendations received
1 person has recommended Zhong
Join now to viewOther similar profiles
Explore more posts
-
Sanchit Narula
The best advice I got on my promotions as a software engineer was from a Principal Architect at Amazon after he told me his story of getting promoted to Principal, and this has stayed with me my whole career. He said, very simply: > Your architecture can be a love letter to your technical brilliance, > but if it does not move customer and company goals, it will not move your level. That hurt a little. Because I recognised myself in it. A few years ago I wanted to design something with every shiny thing in the book: API Gateway, Lambda, DynamoDB, event bus, the works. On paper it was beautiful. In reality it was expensive, slow to ship, and did not match the skills of the team that would maintain it. So we shipped a hybrid architecture instead. Fewer moving parts, more reuse of existing services, cheaper to run, easier to hand over. Over time, here is how my thinking about promotions changed. 1. Start from the business goal, not the tech stack Before talking about Kafka vs SQS or MySQL vs DynamoDB, answer this first: What money, risk, or customer pain will this remove? Promotions track that, not the number of services in your diagram. 2. Ship finished products, not clever demos Leaders remember the feature that went live, stayed stable, and earned or saved real money. Drive the entire lifecycle yourself: design, build, launch, oncall, cleanup, docs. 3. Make the team faster, not just yourself Tools, scripts, templates, better CI, better dashboards. If people quietly say “things move faster when you are on this” your manager already has half a promotion case. 4. Do your share of the unsexy work Oncall, migrations, bugs, incident reviews. Senior folks who avoid this slow the entire org. Senior folks who lean into it become trusted very quickly. 5. Grow people, not just systems Mentor juniors, run design reviews, write clear docs, set standards. When your manager sees that you are already acting like the next level, formal promotion becomes a catch up, not a favour. 6. Be proactive about your promotion, not obsessed with it Have periodic one to ones. Talk about where you want to go and ask what evidence your manager would need to support that. Remember your manager has ten other people to manage. It is your responsibility to make your impact visible. Do not be desperate. Understand the timelines and work within them. As you get close to the boundary, collect the missing data points: impact metrics, incident write ups, design docs, peer feedback. Make the story easy to tell. Most importantly, work for the business, not for promotion-driven development. If your roadmap only exists to tick boxes on a promotion rubric, people can feel it. If your roadmap clearly creates value, saves cost, or reduces risk, promotion becomes the natural side effect. So instead of asking “What new tech can I use this year?” Ask, “Where can I create obvious business impact and leave the system and the people around me better than I found them?”
15 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content