Work at a Startup Expo 2019
These notes were generated by AI and may contain inaccuracies.
Quick round of applause for the companies. It's 2 o'clock; keep on time with a lot ahead. Schedule: opening remarks, Jeff (new president of Y Combinator) on why join YC and work at startups, then first half of pitches (first 20, 2 minutes each), quick intermission, next 20, then around 4 o'clock move back to the room to meet founders and products.
Hi, I'm Ryan. Sent emails with magic links (felt like spam, forgive me). Job at Y Combinator: help founders and YC ecosystem find and work with amazing engineers.
Graduated college, seventh engineer at Salesforce.com (50-person company), worked directly with Marc Benioff, Parker Harris. Built API for developers (Marc's idea), first marketing suite, first internal billing system (twice). Learned more than entire career elsewhere.
Engage with engineers from colleges, startup sign-ups. Love chatting about engineering (8 years at Salesforce, grew with company), engineering management (ran team at Zora, went public last year; most satisfying finding great people, putting in roles to excel). Also products/platforms (PM at Lyft: platform PM, identity PM, IPO PM; partner engineering, developer advocacy). Curious about PM transition? Grousing about PMs? Chat. Startups: work at YC, started YC company, another after; career at startups big/small; MBA at MIT (entrepreneurship/innovation). Email: ryan@yc. Coffee, office, abroad—how can I help?
Jeff, new president of Y Combinator. Loves event: changes trajectories of attendees and YC companies.
Reflected on journey from engineer. Joining startup vs. working for the man. Audience poll: who works at startup? Wants to? Start one?
Why Work at a Startup?
Growth, exciting, risk. Best years: most energetic, creative, free/unencumbered—take risk now (not much risk). Choose carefully, maximize time (ultimate resource).
Future and Change
Back to the Future: 30 years back (1980s), world similar (black/white TV, skateboard). 30 years ago for you: no internet, no mobile, no laptops—lost. 30 years forward: change speeding up. Pair programming with AI (suggests, fixes bugs, structures code) or overseeing AI.
Big Company vs. Startup
Big company: good salary, RSUs, benefits, Apple round building, softball league. Startups: hard, stressful, company could die, long hours, little vacation/support, worthless stock. Uncertain future: no lifetime employment, 5-7 jobs/life, little control/freedom, minimal impact (boring, less innovative). Ryan's Salesforce impact as franchise example.
What Matters
Maximize impact/satisfaction. Most important: team/people—nothing bonds like startup war. Future Slack, stock public at $22B valuation. For founders: no better path than ground-level startup—builds winners, best memory, skills (self-reliance, resilience, communication). Ryan terrible 10 years ago. Tight team, degrees to CEO/customer low.
Jeff's Career Slice
Early 80s California, HP (fur-lined rut). OS group: shell (gone). TCP/IP hacks (HP different TCP/IP, cancelled). Escaped: presence provider for World Wide Web (brought companies online, customer contact, radical; wrote used software). Next startup failed (wrong co-founder): wrote code, customer support (Emacs-based person lookup/free web page—should've been product), product person. Then best team: Dave Nakayama, Anton Olivtani, Larry Dreeben, Brian Woods—built Rocket Mail (became Yahoo Mail, late 90s). Acquired late-career (wife/kid too)—not impossible.
How to Choose a Startup
Check workatastartup.com, Startup Expo. Find problem you care about, team you like. Collison true laughter: hire if happy finding them only one in office Sunday (you'll work Sundays). Search mission to sacrifice for. Research, proactive, referrals (tight community). Variety sizes/sectors. Startups hard to resist good person who knows/wants in.
Why YC
YC since 2005 (14th year), funded 2000+ companies, 4000+ alumni founders (largest network), $150B+ combined valuation. 100+ funded today >$100M valuation. Top 100 created nearly 30,000 jobs. Great path; gives skills/satisfaction for life/career.
Switch slide. Email magic link (raise hands). Open email, click link → directory (or URL/QR code). Directory: companies in pitch order, like button (sends message to founders, bookmark). Countdown to start pitches.
Vijay, founder/CEO AirRx Health: remote monitoring for chronic lung disease patients. Stanford MD, two healthcare IPOs. Team: three YC founders. Chronic lung disease → #1 hospitalizations soon. Software checks elderly daily, helps doctors catch/treat early (keep healthy/out of hospital). Live, revenue-generating, stealth. Oversubscribed round: Floodgate, Village Global, Acme Capital. Tech stack to save lives. Growth: Silicon Valley mentors, development stipends, diverse team.
Tane, co-founder/CEO Thalys: computer vision + microfluidics for instant point-of-care blood monitoring. Device: microfluidic channel. Trained 10s convolutional NNs to segment/classify 100s cell types → complete blood count. Video demo: finger prick → test strip → device → CBC in minutes. Deployed in 100s clinics, tons revenue. Major FDA clearance: first single-drop neutropenia test (10M immunocompromised: cancer, immunosuppressive drugs), first deep-learning FDA-cleared test. Foundation for cancer/allergic tests. Contracts: largest pharmas, $100Ms. Deployed 1000s patients/100s clinics. Raised $20M+: Sequoia, Nvidia, YC.
20% female founders sexually assaulted/coerced by investor; <20% report (don't hurt company future). Victims forward to protect others (90% repeat offenders). Anja Norma Jean, CTO Callisto: sexual misconduct reporting software. Two+ victims same perpetrator → connect, options. Process: time-stamped encrypted record (unique perpetrator ID, escrow—nobody sees unless match). Match → legal options counselor (attorney, privilege). Launched 4 years ago colleges; serve every survivor. Bleeding-edge security: stop violence, protect privacy/civil liberties (victim/accused). Stack: Node/Angular. Expertise: advanced cryptography, trauma-informed design. Backed: Greylock, Skoll Foundation, Google.org. Join brave/brilliant engineers.
Jonathan, co-founder/CTO Checkr: background checks (5 years). Mission: fairness/transparency. 1.5M checks/month, growing. Clients: Uber/Lyft (drivers), on-demand/mainstream. 100-person team, doubling yearly (new teams, career growth). Profitable. Engineering: identity matching (core accuracy), ML. Most backend, reliable integrations/data sources. Impact millions users/1000s customers/large corps.
Like video games? League of Legends, PUBG, Fortnite, Counter-Strike, Dota: billion-dollar, originally individual modders (no pro chance without big recruit). Dreamcraft: platform for millions create/sell games. Needs: simulation (render worlds/power gameplay), editor (create content), release platform (distribute/sell). Tech stack: figure out. Fun: games, help people (satisfying). Video of possible tech.
Julia, CEO/co-founder Dinette Technologies: mobile authentication, credit card scanning SDK. Founders: ex-Uber (built Uber app scanner), first engineer ex-WhatsApp early Android. Optimize CV model on-device. Launched end-January: enterprise customers, seed from exceptional investors. Stack: deep learning (Python backend, TensorFlow Lite frontend), mobile (server-side Swift backend, Swift/Java frontend). Join if: mission (safer internet, tie identities to crimes), maximize autonomy/impact (<10 people, founding team outsized culture/future), intersection real-time mobile fraud + deep learning edge. Unmarked table near blue truck; [email protected].
Danielle, founder Faire: wholesale marketplace for independent retailers. 2.5 years, $100M+ from Sequoia, YC Growth, Lightspeed, Khosla—one fastest YC. Future local: help small retailers vs. Amazon/Walmart (curated, communities thrive). Tech: Kotlin/Java, Swift iOS, Kotlin Android, web. Marketplaces challenging (two customers). Practices: automated testing, deploy often, readable code (learnings from Cash App first engineers). Data important (chief data officer + CTO founders). Test/deploy quickly.
Mike, cofounder GiveCampus: platform helps schools raise money. Education important; U.S. much money runs
Fundamentally, education is important, but in the U.S., much of the money for schools comes from private donors. Many graduates receive phone calls or direct mail for donations, but in 2019, it should be easy to give online transparently, knowing where the money goes, and hopefully social. They focus exclusively on schools to build the best product. Team includes former Facebook, Amazon, and Intel engineers, and on the business side, former school fundraisers with empathy for end users. Current team of 28 is over 50% women, hoping to expand to 50 by year-end. Profitable since early 2016. Tech stack: Ruby on Rails, Postgres, Heroku; front-end vanilla JS and React. Headquartered in DC, hiring everything in DC and in San Francisco for a tech lead or senior role.
David, CEO and co-founder of Glide, makes apps from spreadsheets. Demo: Choose new app, pick a sheet from Google Drive; Glide creates a default app configurable with components that bind to spreadsheet data and sync bi-directionally. Customize layout, theme, add media like photos, then share with employees. Example: Bakery stays on top of inventory. Launched four months ago, featured in last YC batch, already thousands of non-technical software developers building and distributing apps monthly for everything spreadsheet-based. Targets 'dark apps'—internal private mobile apps enterprises build without app stores. Favorite story: Largest real estate developer in St. Louis, Missouri ran on 7 spreadsheets; employees carried laptops. One non-technical employee built a Glide app opened 30 times a day; now no laptops needed, powers their business. Ambition: Convert billion spreadsheet users into productive software developers. Seven people, looking for eighth; more than half the company here. Investors/advisors include CEOs of GitHub and Figma. Tech stack: TypeScript, React, Firebase. Half designers, half engineers; booth has printouts of upcoming designs for front-end engineers.
Greg, co-founder and CEO of Goodly. Student loan repayment as employee benefit, named hottest by Forbes. Functions like 401k: Employers pay off employees' student loans. Growing rapidly; 4% of U.S. companies offered it when submitted, now 8% (doubled in 12 months), projected 32% in next two years. Employees make regular payments; Goodly software lets employers make secondary payment directly to debt, helping average employee pay off 30% faster. Hiring first full stack engineer; currently CTO (co-founder) heads engineering. Team: Sales, growth, CTO. Backed by Y Combinator and Norwest (Jeff Crowe, Forbes Midas list 6 years). Recent press coverage as benefit grows.
Ning, co-founder and CTO of HealthTrippa. Mission: Help Americans access healthcare via affordable ACA marketplace insurance with subsidies. Enrolled 1.8 million people (7% of ACA enrollment last year), saved them $6 billion on premiums. Generates >$10M revenue/year, highly profitable, up 3x yearly. Tech: Ruby on Rails (API), React, Python/R for data analysis/science, Elasticsearch/MapReduce for back-end data products. Impact: Serve $21K/year household income, save them $9K/year on insurance. Unique: First with direct API to healthcare.gov backend. Fill outreach gap amid rollbacks; primary integrator for insurers, employers, brokers. Market: 45M uninsured Americans. Work: Full-stack with 4 engineers, own products. Projects: Health insurance recommendation engine, government API integration passing FedRAMP audit (now government contractor-eligible), architect for 20x traffic spikes. Work in San Francisco, Sacramento, or remote; family-friendly, flexible hours.
Paul, co-founder and CTO of Human Interest. Helps everyone save for retirement via software. Product: 401k for startups/small businesses; automates paperwork/compliance, syncs with payroll, single-click investing like Wealthfront in a 401k. Mission: Only 20% of U.S. small businesses offer 401k (primary retirement savings vehicle); tech enables it for 5-1000 person companies. YC 2015; co-founder Roger (third successful startup); Paul interned at Justin.tv (pre-Twitch). From kitchen to 65 people in downtown San Francisco; helping 1000+ startups/small businesses, 75% never offered 401k before. Hiring back-end leaning engineers who work with PMs, sales, support to define product—great for aspiring founders. Tech: JavaScript-heavy, full-stack TypeScript, Node.js, Postgres backend, React/GraphQL front/middle. Example project: 3-person team rewriting core accounting system. Culture: Smart, humble, hard-working (college dropouts to MIT). Investor/former VC now CEO; Series B imminent; next: IRA, HSA.
Samir Magani, co-founder and CEO of Instawork. Mobile app connects gig workers with hospitality (hotels, restaurants, catering). Helps people get jobs, increasing employment. Small/blue-collar/hospitality lack LinkedIn; rely on Craigslist/signs. Built API for labor: Request like Uber (bartender/dishwasher); handles curation, matching, billing, insurance. Raised ~$30M including Series A from Bill Gurley (Benchmark); just raised/announced Series B due to growth. Hiring engineers/data scientists. Tech: Modern stack, serverless for scalable infrastructure, in-house Hyperview framework for faster mobile dev. Values: Respect team/user/craft; empathy (all employees work a gig; founders in bistro attire to be close to users).
Tariq, CEO/co-founder of Kashi. Coffee is global betting site to bet on anything with anyone (Brexit by month-end, Oscars winner). Vision: Capitalize on opinions. No U.S. prediction markets; on track to be only legal one, untapped market/no competition. Hiring talented engineer for back-end infrastructure (state-of-the-art trading exchange), seamless UI for trading newbies, ML models on trade data for market surveillance. Luan and Tariq: MIT CS/math, tech/trading experience. Super small, hard-working team.
David, co-founder of Keeper. AI bookkeeper for 50M U.S. gig workers (Uber drivers, freelancers, realtors) overpaying taxes ($2K+/year full-time). Gig workers track expenses (phone, travel) poorly via shoebox receipts/spreadsheets. Solution: Link bank account; use work profile + ML to find expenses, notify via text (chat interface). Tech: React, Node, Python for data science. YC recent batch. Full-time: David (5 years quant trading/data science), Paul (Harvard, Square/Amplitude PM). Hiring founding engineer to drive engineering decisions (founders non-strictly SWE).
Lea, CEO of Lob.com (YC summer 2013). Vision: Make world programmable—connect APIs to offline world. Products: Address verification API (powers Amazon checkout suggestions globally); Print & Mail API (e.g., Comcast statements automated). 6 years old, $80M pace this year; powers biggest companies; 61st largest YC by valuation, top 25 by year-end possible. Tech: Migrating from Node to Go/Elixir for scale (100M API requests/day). Challenges: Parse single-line addresses into components for deliverability (hard outside U.S./USPS in 280+ countries); optimize mail routing (1/4 U.S. households receive Lob mail). Culture: Build for 30+ year tenure; progressive, creative (beyond Silicon Valley norms); love dogs.
Chris Winn, co-founder of LogDNA. Engineering product for DevOps: Helps debug/monitor production issues—first look when shit hits the fan. Aggregates server/app logs into one platform. 3,000+ customers (many in booth), $35M funding. Post-YC: Near death, pivoted with $17K left; now $35M in bank. Investors: YC OG Gary Tan, Alexis Ohanian, Emergence Capital (Zoom IPO). Tech: Node, Vue.js, Kubernetes, Rust, Elastic. Customers: Reddit, Instacart, Lime scooters; powers IBM Cloud global logging (incl. Watson). Leaders from Facebook, Heroku, Nvidia, eBay. Values (HOT PIES): Honesty, Open communication, Trust, Passion, Ideas (no box), Execution. Tripled ARR in 6 months (Series B); built new Kafka 'Buzzsaw' (open source soon). VP Eng: #200-7000 at Facebook.
Erica, co-founder of Modern Health. Crunchbase #1 startup 2019. First mental health platform for companies; team of 20 from gaming/healthcare/security. Depression #1 global disability cause; proactive solution for stress/depression/anxiety regardless of employer benefits. Tech: Python/Django, React, GraphQL; DevOps with data-driven monitoring/alerting; engineers own features + infrastructure. Tripled in 3 quarters; $11M+ from Kleiner Perkins, YC, Jared Leto. Live what we sell: Employee well-being #1 value.
New Front Insurance in commercial insurance industry; companies spend over $500 billion/year on insurance and it's
Foundational to our economy, every business needs insurance, yet it provides a terrible client experience. Building a full-stack startup: software and technology platform to support operations and provide service to clients. Crazy growth, white-knuckle growth: started 2018 with 135 clients, now almost 3,000 clients.
Technology goal: deliver amazing experience to clients when they transact and manage insurance. Back end: TypeScript and Go microservices. Front end: React plus hooks. Infrastructure: Heroku, CircleCI, Datadog for telemetry.
19-person engineering team, growing with four people starting next few weeks. Diverse group: people from insurance industry, mid-career (speaker has young kids), parents and grandparents, 30 years in insurance. Draws from two worlds. Culturally work directly with users, sit in same room, ship to solve real problems in live operational environment, see people using software same day.
Measurable visible successes: well instrumented, metrics business, commercial insurance broker well understood. Ship software, impact metrics, see and point to success.
Dureena, one of founders of OpenPhone. Gives business phone number in app on cell phone, building modern business phone. Current solutions outdated. Launched one year ago, YC Summer 18 batch, 4,000 paying customers. Need help to build product quickly.
Stack: mobile apps iOS and Android, working on web app. Challenging problems, product used daily by people relying on it for business. Solving big problem in massive market. Small but mighty team, recently closed seed round. Build future of communications, talk at OpenPhone table.
Nick, co-founder of Outschool, previously high school teacher and first engineer at Airbnb. Marketplace of live online classes for kids. Different from other EdTech: consumer marketplace bringing together parents, teachers, learners. Format: live online small group video chat classes, make learning social and engaging.
Product: broad, marketplace and communication tools for teachers to run classes and communicate with kids. Modern web app tech stack, build on other tools like Zoom API for video chat. Currently 5 people, hiring 5 more engineers by end of year.
Reasons to join:
- Great team, experienced founders focused on building intentionally.
- Immediate positive impact: product used by 1,000 teachers and 30,000 families, teachers earning full-time living, hear from users how it makes difference.
- Great business: product people love, 5x growth last year, sustainable model, network effects, Series A from Union Square Ventures last fall, capital to grow team.
Biggest challenge: hiring engineers.
Wei, CTO and co-founder at Patch Medical, Australian. Come work if you deeply care about people, obsessed about solving difficult problems, optimize career for growth and new experiences, make disproportionate impact and have disproportionate responsibilities.
Triaging and preventing readmissions due to sepsis in discharged patients using wearables and deep learning. Sepsis: body's dysregulated response to infection, affects 6x more Americans than breast cancer each year. Personal: co-founder had 18 episodes.
Built and validated algorithm on 48,000 patient records, beats standard of care on area under curve, sensitivity, specificity, positive predictive value. Highly technical team, backed by key medical advisors. Hiring machine learning and software engineers for FDA approval system. Recent regulation changes make healthcare ready for AI.
Vision: no one should die from preventable sepsis episodes, address with wearable monitoring and deep learning models to predict.
Lila Avery, founder of Pull Request. Build code review as a service: network of on-demand reviewers with automation for high code quality. Everyone commits bugs, help teammates. Process 3 million lines of code, tag to categorize bugs, causes, solutions. Building AI for this.
Work with GitLab, GitHub, Bitbucket. Stack mostly Go and React, review any language: Clojure to Elm to C++ on Apache. Build future of development, hiring for everything, passionate about engineering, work with thousands of world-class engineers.
Gregory Coburger, CEO of ReadMe. Help companies build amazing developer experience for APIs. Most have used APIs. Favorites like Scale, Stripe, Twilio have beautiful interactive documentation, bring to thousands of API companies.
Tech stack: Node, Angular, React. Handling big data from API logs for understanding and debugging.
Why join:
- Great benefits, good pay.
- Quarterly offsites entire team: Hawaii, Austin, Chicago for product roadmaps.
Interview process:
- Website with all info: who's interviewing, what to expect, prepare.
- Bring own project, not assigned tech project, see excitement and work on known codebase.
Team over 20, hiring front-end, back-end, full stack, data, back-end databases, dev evangelists, support engineers. Growing quickly.
Adam Mackenzie, CTO and co-founder of Rescale. Companies with heavy scientific computing have fixed supercomputer capacity: months to install, costly, idle low times, queues peak times. Rent elastic on-demand supercomputer ready in minutes, customized for workload.
Build on major public cloud providers, top commercial, open source, private simulation software. Bring products to market faster.
Stack: React front-end, Python Django web tier, Postgres, Redis, Java Dropwizard services for cluster orchestration across clouds.
Why work: great team HQ in SF, 100 people globally. Exciting customers: Nissan crash test analysis, Boom (YC, supersonic travel). Hiring all types: Engineers, DevOps, front-end, back-end, data.
Jacob, CEO co-founder RevenueCat, YC last year. In-app subscriptions made easy: subscriptions on App Store or Google Play. Help developers make more money with apps. SDK and back-end out of box, add subscriptions few lines code, up in days. Dashboards, integrations, no worry scaling business, focus on users.
Help with price testing, LTV analysis. Dashboard in React. Doing well, great month recently.
Stack: Python, Postgres, React, AWS, Heroku. Simple to avoid complexity. Serving 50,000 requests/min, peaks 80k, get sweaty. Hiring vacuum people, databases at scale (previous biggest 3k/min). Engineering driven culture, all engineers building for engineers. Love tests. Developers as customers, hiring back-end, developer support (buried in Intercoms).
Parker Conrad, co-founder and CEO of Rippling. HR and IT software for small businesses. Analogy: wouldn't set up 100 separate databases for revenue/customer data. Employee data tracked in ~100 disconnected systems (usernames, passwords, department, manager). Causes pain: onboarding hard, admin work.
Solution: one underlying system to hire, change, terminate employees, propagates to everything. Rippling does that. NPS high 60s. Month-over-month growth rates up and to right for latest financing. Raised Series A from Kleiner Perkins, ~$100M total.
Hiring lots of roles.
Alex, CEO and founder of Scale. Vision: accelerate AI applications development. AI appropriately hyped: if software eating world, AI eating software. Bet on AI like Internet in 80s, full impact extremely large over decades.
Supervised deep learning (exciting AI results) requires incredible data amounts. Bottleneck: massive data needed. Engineering change fundamental: traditional software improves by writing more code (loop: write, deploy, fix). AI development different way to improve.
In AI development and machine learning, a new way to improve system performance is by feeding more data to the algorithm. Machine learning teams at companies like Apple and Google follow a loop: label data, retrain model, evaluate, feed more data. Scale AI built an end-to-end API-driven platform for ground truth data, including a dashboard and toolset for labeling data, an API to send data to them and receive it back, and a custom dashboard for customers to oversee it. Behind the scenes, a network of labelers handles the labeling, accelerated by machine learning.
They work with customers like OpenAI, Waymo, Cruise, Lyft. Team grew by more than three times in the past year. Raised more than $20 million from Accel, Index, Greg Brockman, Not Boring, etc. Check them out at their booth.
Segment is an analytics platform that helps companies collect data on who the user is and what they're doing on the site, integrating with 250 tools like Google Analytics, Marketo, Kissmetrics, Optimizely, BigQuery. It has extensive customer data infrastructure with many services and products, similar to AWS.
Personas launched in 2017, to market in 2018. Started with five engineers in a product-market fit cycle working closely with customers; now a $10 million business two years in. Launch about two products per year. 19,000 customers across startups and enterprises. Build data pipelines, handling average 250,000 messages per second; peaks at 750,000 messages/second during Super Bowl or Indian Premier League. Within an order of magnitude of world's largest APIs like DynamoDB and Google Analytics. Do cool visualizations with React front-end stack. Open source with 500 repos, 50,000 stars.
Hiring: Personas engineering managers, full-stack engineers.
Simply Insured is a 100% online way to buy health insurance for small businesses across the US. Instantly quote every health plan from every carrier, 100% online application and management.
Customer story: Woman in ER at 8 p.m. Saturday because husband had stroke, no ID card. Couldn't call insurance company. As Simply Insured customer, used mobile ID cards product to log in and access info instantly. Desk person said coverage invalid despite paying $1,000+/month. Logged onto site, called support; navigated to confirm coverage, enabling life-saving surgery.
Mission: Eliminate fear from health insurance. Software for quoting, applications makes doctor's office/ER experience amazing.
Tech stack: React + Redux front-end, Ruby on Rails backend, AWS. Signed up Intuit and Square, largest small business companies, distributing platform. Tripled team in last year.
As work moves to cloud apps like GitHub, Asana, Trello, Google Docs, human-computer interaction regresses; can't search or create folders across work.
Slapdash builds an operating system for cloud work across these apps. Do 80% of cloud app tasks in one interface: search own/company work, close tasks, tickets, merge PRs.
Holy grail in productivity: integration. Winner must be neutral with fresh approach. Build giant graph of data across cloud apps/company apps. Framework mutates graph on changes in apps. Database access layer with batching/coalescing for low-latency desktop-like experience. Consumer-grade stack: TypeScript client/server, GraphQL, React.
Why join: Work on hard problems with tangible product you'll use. See everything built from ground up (three founders). CEO Ivan built Facebook business unit generating $5-7 billion. CTO Dima sold three companies, rebuilt WhatsApp infrastructure for 30 million requests/second. Third YC company for speaker to take from zero to $10M revenue.
Sorceress builds AI to replace human recruiters. Dogfooded product to build amazing team.
Team highlights: Co-founder was Dropbox's first chief of staff, paid through MIT via algorithmic trading. Speaker published ML papers, sold company. Investor led Box. Engineer Abe CTO of another YC company. Engineer created OpenAI internship program. Engineer built state-of-the-art Python profiling system as side project. Engineer featured in deep work book. Engineer Tina had scholarship for ML research at Recker Center.
Care about learning/growth: Teach ML or engineering management, weekly talks, Paper Club, OpenAI advisors, pay for coaching/conferences/classes/tools.
Make world better: Help companies curing cancer, equalize opportunity, reduce hiring bias, help people find happy jobs.
Customers with $2.5B+ annual revenue. Tech stack: Python, TypeScript, React, PyTorch. Raised $10M Series A two months ago. Double team next year. Benefits, free meals.
Computer vision revolution: Computers seeing as well as people changes everything, like mobile/Internet.
Not autonomous vehicles (10-20 years off). Massive application: Autonomous checkout in real stores. E-commerce <10% of spending; real stores $25T/year, $200B on cashiering. Billions of human hours waiting in lines.
Complex but deployable now: Mistakes mean free ketchup, no one dies. Retailers/shoppers okay. Amazon deployed ~10 stores last six months; they deployed first store six months ago, multiple this year.
Challenges: Real-time from cameras only. Identify every product through occlusion/complexity. Visual tracking: Identify people across cameras in crowds without facial recognition (unsolved problem).
Store in San Francisco. Hiring for ML, production engineering, high-performance computing, front-end/back-end.
Funkable enables anyone to create beautiful/powerful mobile apps for personal/family/team/world use, not for app stores.
Drag-and-drop UI design, visual programming for minimal coding.
Stack: React front-end, Node/Express backend, React Native mobile, GraphQL, Redis, MongoDB, Google Cloud/Kubernetes.
Hiring full-stack engineers excited about mission. Committed to diversity.
Why join: So your mom understands what you do. Turn passive consumers into creators to understand tech work.
Theorum in fintech lending solves 2007 financial crisis issue: Loans sold by originator who takes fee, not risk; prices to maximize volume, not right price. Leads to bad loans, defaults, investor losses.
Embed APIs in lenders' systems: Send all loan data (not just FICO), get pricing. Guaranteed buyer if priced to their rate before borrower offer.
Started five years ago, fund $60M loans/month; grow to $200M in 18 months. Work with major Silicon Valley online lenders, expanding to traditional.
Goal: Safer financial system with unmanipulated loans for consistent capital.
Stack: Heavy ML for underwriting, data/infra engineers for <100ms offers. Small team (23 people), $15M+ revenue. Tech startup + academic research on deep problems.
VoiceOps helps call centers understand customer conversations. Target: Sprawling sales floors making thousands of calls/hour.
Problem: Executives can't see if reps say right things to close deals.
Process: Take call recordings, transcribe to text, analyze sentence-by-sentence for key events, dashboard with structured data to pinpoint issues.
Traction: Raised $11M, customers like Square, 25% month-over-month growth.
Engineering: Join as engineer #3, own lots, big impact. Frontier problems: Structure messy dialogue accurately, scale/optimize pipelines.
Team: Experience-agnostic; seek smart/tenacious. Tight-knit, trust-based, intense (not work-life balance), rewarding mission.
Valensa builds/operates long-range, high-payload UAVs for point-to-point time-sensitive deliveries. YC W17, backed by Lightspeed, raised $25M+, 30+ engineers/operators in Bay Area.
Customers: US DoD to Caterpillar, Merck. Post-revenue.
Applications: Defense (surveillance, critical supplies), commercial (spare parts to cut downtime), medical (vaccinations, blood).
Projects: Bahamas, Ireland, Puerto Rico, Africa, SE Asia.
Vision: Autonomous aerial mobility starting with cargo, change transportation for goods/people.
Portfolio: 10-200 lbs payload, 60-1,000+ miles.
Robotics-first company building flying robots. Openings on website; at booth.
Bloo (remember Zyphe) makes world better without Facebook ads. Help brands bypass Facebook to build community.
YC W18, raised Series A, on track $6M ARR. Brands: Nike, Lyft, Dior, Kellogg's.
Stack: Python, React Native, TensorFlow ML.
Challenges: Four pillars, roadmap with interest-based communities, front-end/mobile APIs, data/ML (filed computer vision patent, predictive analytics, recommenders).
Team: 12 in product/engineering, keep headcount low for impact. Hiring: 3 roles (front-end, back-end, data science), product manager, UX designer. Reese's Pieces in back.
Day over. If thinking of leaving, come for boba. Mobile in back. Companies excited to meet. Head back over.
Keep Y Combinator in your library
Save the videos and channels worth coming back to, and find them again in one place.
