<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Tom Hauburger on Medium]]></title>
        <description><![CDATA[Stories by Tom Hauburger on Medium]]></description>
        <link>https://medium.com/@thauburger?source=rss-7b0f02ef02e3------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*LPXaYWdyTsuDlPvIpmXM-Q.jpeg</url>
            <title>Stories by Tom Hauburger on Medium</title>
            <link>https://medium.com/@thauburger?source=rss-7b0f02ef02e3------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 09:49:03 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@thauburger/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[A Safety-First OS for Voyage's Driverless Technology]]></title>
            <link>https://medium.com/voyage-news/a-safety-first-os-for-voyages-driverless-technology-f87da4c363d3?source=rss-7b0f02ef02e3------2</link>
            <guid isPermaLink="false">https://medium.com/p/f87da4c363d3</guid>
            <category><![CDATA[autonomous-cars]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[self-driving-cars]]></category>
            <dc:creator><![CDATA[Tom Hauburger]]></dc:creator>
            <pubDate>Tue, 19 Jun 2018 15:02:50 GMT</pubDate>
            <atom:updated>2018-06-19T15:02:50.424Z</atom:updated>
            <content:encoded><![CDATA[<h3>A Safety-First OS for <a href="https://voyage.auto/">Voyage</a>’s Driverless Technology</h3><h4>Announcing <a href="https://renovo.auto/">Renovo</a> as the development platform for our <a href="https://news.voyage.auto/introducing-the-voyage-g2-autonomous-vehicle-5e15cca399b5">G2 autonomous fleet</a></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tkdSzCbh8LFdj5erhMDmjA.jpeg" /></figure><p>At <a href="https://voyage.auto/">Voyage</a>, we fundamentally believe that the autonomous vehicle industry is at an inflection point. The number of talented companies focusing on specific verticals in the AV industry has created an ecosystem where — for the first time — a safe and robust autonomy stack can be built leveraging the expertise of partners.</p><p>Today, we’re excited to announce one such partnership. <strong>Voyage has selected </strong><a href="https://renovo.auto/"><strong>Renovo’s AWare</strong></a><strong> as the platform for developing our autonomous driving technology.</strong> Renovo shares our belief that the future of autonomy requires the ability to easily and safely integrate the best functionality on the market. AWare is built from the ground-up with Level 4, driverless applications in mind. From real-time execution of safety-critical operations, to a growing ecosystem of drop-in partner integrations, AWare lets us focus on what we do best: building autonomous driving technology for our communities.</p><p><em>What do we mean by “platform”?</em> Renovo delivers a critical layer between the vehicle and Voyage’s autonomous driving software. It provides programmatic access to the vehicle’s sensors, safety systems, and motion control mechanisms. Popular middleware solutions like <a href="http://www.ros.org/">ROS</a> are great for prototyping autonomous systems, but start hitting functional limits when building fully driverless systems. Renovo not only delivers enhanced functionality designed specifically for Level 4 applications, but also exposes a safety architecture and partner ecosystem that provide us with confidence that our autonomous driving technology will run safely, reliably, and with the best tools on the market.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wDcb61YYxhv3za_N54PSuw.jpeg" /></figure><h3>Safety-First Architecture</h3><p>At Voyage, our goal is to completely remove the driver from our fleet vehicles. In doing so, our top focus is always on passenger safety. We needed a platform solution that shared our safety-first philosophy.</p><p><strong>Renovo was designed from the ground-up to support truly driverless use cases.</strong> Some of the most significant safety features that differentiate AWare include:</p><ul><li>Real-time execution of safety-critical procedures</li><li>Hardware-level vehicle state machines</li><li>Quality of Service (QoS) primitives and configurable resource limits per service</li><li>Architectural flexibility to support redundancies in power and compute</li><li>System diagnostics and configurable fault handling</li><li>Support for tele-operations and remote braking functionality</li></ul><p>In addition to these features, the Renovo team has extensive industry experience in delivering <a href="https://news.stanford.edu/2015/10/20/marty-autonomous-delorean-102015/">safe, high-performance electric and autonomous systems</a>, making them a strong validation resource as we implement our own system-level safety redundancies.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Hd7t6-JezPgRUUD8JIuxdA.jpeg" /></figure><p>The shared focus on safety is so strong, <strong>Renovo has agreed to join Voyage in the development of our </strong><a href="http://oas.voyage.auto/"><strong>Open Autonomous Safety (OAS)</strong></a><strong> initiative.</strong> We’re excited to bring them on as an official partner!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5F2YSTiOzRcEDqEhAGMDJg.jpeg" /><figcaption>Driving at The Villages, San Jose</figcaption></figure><h3><strong>Production-Ready Middleware</strong></h3><p>Building an autonomous vehicle is hard. AWare simplifies development by providing high-quality, production-ready middleware addressing common infrastructure needs of autonomous systems.</p><p>Specifically, AWare provides:</p><ul><li>Robust intra-service messaging</li><li>Automatic, configurable telemetry, logging, and local storage</li><li>Vendor-agnostic abstractions for LIDAR and camera sensors (<a href="http://www.velodynelidar.com/"><em>Velodyne</em></a><em> support is great!)</em></li><li>Vehicle-specific deployment configurations</li><li>A flexible, time-series query language for on-vehicle data (e.g. camera frames or telemetry events)</li></ul><p>Renovo has also done fantastic work to help us intelligently off-load and store data when our vehicles return to base each night. Their data layer supports a query interface that ensures we capture the most important data from the day, without unnecessarily waiting to off-load terabytes of redundant data from each vehicle.</p><h3>Growing Partner Ecosystem</h3><p>One of our favorite parts of Renovo’s approach is the belief in an open partner ecosystem. In many ways, Renovo has adopted the the successful <a href="https://aws.amazon.com/">AWS</a> model: <em>“we provide the infrastructure, you choose the tools”</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*wePGsZFkxGBGsnMd4-pgXg.gif" /><figcaption>The Velodyne VLS-128</figcaption></figure><p>As a startup, one of Voyage’s most expensive investments is in one-off partner integration. Renovo helps address this through its growing catalog of drop-in integrations with industry partners. From mapping and telemetry, to sensor drivers and tele-operations, Renovo is actively building an ecosystem that allows any AWare customer to leverage a host of off-the-shelf integrations to enhance their autonomous service.</p><p>This model fundamentally makes it easier for Voyage to experiment with different industry partners and select the best functionality for our use case.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hxZHHzhVXFFzlvcTfFcWgQ.png" /></figure><p><strong>We’re extremely excited to have Renovo as a core strategic partner. Voyage is actively integrating AWare, and will be rolling it into production as part of our </strong><a href="https://news.voyage.auto/introducing-the-voyage-g2-autonomous-vehicle-5e15cca399b5"><strong>G2 vehicle</strong></a><strong> release.</strong> Renovo’s safety-first design is an important component of our driverless taxi service, and the relationship is a great example of partnering with the industry’s best to accelerate our path toward a truly driverless taxi service.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C2NkTAr6y7vrzTOPuEhcmw.gif" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f87da4c363d3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/voyage-news/a-safety-first-os-for-voyages-driverless-technology-f87da4c363d3">A Safety-First OS for Voyage&#39;s Driverless Technology</a> was originally published in <a href="https://medium.com/voyage-news">Voyage</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[26 Hours]]></title>
            <link>https://medium.com/@thauburger/26-hours-3c4a820c8d4?source=rss-7b0f02ef02e3------2</link>
            <guid isPermaLink="false">https://medium.com/p/3c4a820c8d4</guid>
            <dc:creator><![CDATA[Tom Hauburger]]></dc:creator>
            <pubDate>Mon, 28 Apr 2014 18:47:25 GMT</pubDate>
            <atom:updated>2014-09-26T05:44:34.230Z</atom:updated>
            <content:encoded><![CDATA[<h4>How I met with 26 startups for my birthday, and what we learned</h4><p>For my birthday this year, I decided to give away 26, one-hour sessions to 26 different startups. A number of people have since asked about the project, so I put together a quick recap below. Thanks to all the startups who reached out and made this such a cool experience.</p><h4>The Offer</h4><p>The morning of my 26th birthday, I threw together a [<a href="http://hours.tomhauburger.com/">simple site</a>], and posted an offer to Hacker News. The terms were simple: send me an email, and we can spend an hour working on whatever you think would be most helpful to your startup.</p><p>I hadn’t thought much about the idea before that morning, and had no idea what to expect. I posted the offer around 3pm PST that day, and went out for a birthday dinner.</p><p>Over the next 48 hours, I had more than <strong>120 startups</strong> reach out for one of the 26 spots. I spent the weekend replying to every email, and setting up time with 26 of the companies.</p><h4>The Responses</h4><p>The responses I received were all over the map. Everything from established startups I had used myself, to first-time entrepreneurs with little more than an idea.</p><p>Over the next three weeks, I held meetings over Google Hangouts with each company that scheduled a slot. I had offered to work on <em>anything</em>, but the discussions basically fell into 4 buckets:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/615/1*846Qqrtcs_8V0fSHpQcwTQ.png" /></figure><p>The number of companies wanting to talk specifically about product launches surprised me a little. In retrospect, a company who is already preparing for / thinking about a launch might be a more likely to respond to an immediate offer for advice and validation.</p><p>I was also surprised at how <em>few</em> companies asked about fundraising. I was expecting there to be a little more “Hey, you were in YC, right? How do we get them to fund us? Can you intro us to investors?”.</p><p>Fortunately, I got very few of those questions. The folks wanting to talk about fundraising had legitimate draft decks, great questions, and were looking for feedback. I was really pleased with how those discussions went, and was happy to hear that 3 of those companies have raised money since.</p><p>4 of the companies had no demo or live product at the time we met.</p><p>4 of the companies had raised more than $1M in funding.</p><p>The other 18 companies had a live or demo-able product, with less than $1M in funding.</p><h4><strong>The Sessions</strong></h4><p>I really enjoyed meeting with the startups. I kept my own personal “score”, and felt helpful in 24 of the 26 meetings. Hopefully, the startups felt similarly.</p><p>A few common themes emerged during the discussions, and I’ve captured them here along with some of my thinking and advice.</p><h4>1: The mythical “launch”</h4><p>A large number of startups wanted to talk about launching their product. In general, I observed that many of these startups had created a mental barrier in defining the <em>launch</em> as some make-or-break moment for the company.</p><p>It’s easy to place an artificial importance on “launching” a product. As entrepreneurs, we tend to be over-achievers, and sharing an unfinished product with someone is hard. It’s hard because we’re consciously making the decision to open ourselves up to criticism.</p><p>At the same time, we know it’s impossible to launch a perfect product. There’s always an excuse to push a launch out two more days, a week, a month. But if we’re honest with ourselves, we’re often just delaying the potential discomfort of criticism.</p><p>At [<a href="http://letsembark.com/">Embark</a>], we “launched” more times than I can count. Sometimes, that meant a Tech Crunch article and lots of new downloads. Other times, it meant releasing support for a new city on the App Store, and not having any meaningful change in users.</p><p>Some launches are better than others, and you get better at understanding which channels of communication work, which things you should measure, etc. But at the end of the day, a launch is whatever you make it.</p><p>As a young startup, there’s no reason you can’t launch one day, change the title of your Hacker News post, and re-launch the next day. Launching never has to stop — and it gives you the most valuable capital you can have: <strong>actual user feedback.</strong></p><p>For the vast majority of startups, <strong>there is no “mythical launch” that makes or breaks the company</strong>. I like to think of launching as sharing my product with other people and asking for a reality check. If you can do that, it becomes less intimidating.</p><p>If you have something that can provide someone with net new value, ship it. See what happens. It’s almost always good for the product.</p><h4>2: Pitch Decks</h4><p>I saw a number of pitch decks during the sessions, and they all faced the same challenge that <em>every other </em>pitch deck faces: too much information.</p><p>When you first start working on a pitch, the natural tendency is to write down everything cool about your product. The problem is that most founders think that everything about their product is cool — which is great, but leads to <em>really</em> long decks.</p><p>I encourage startups to go through the process of writing that really long pitch deck first, with the understanding that it’s <strong>not the final output</strong>.</p><p>All the information you write down in that first pass is valuable thinking. It’s good to get on paper. But 95% of it exceeds what is necessary or consumable for your audience.</p><p>Look at the first pass of your deck, and put it through the “Demo Day” test. Assume your audience is sitting in a hot, crowded room after seeing 50 other pitches.</p><p><strong>In three minutes, does your deck communicate what you’re building, why it’s going to be huge, and why your team is awesome?</strong></p><p>If so, you’ve got a fantastic deck. If not, work on getting it there.</p><h4>3: Outsourcing Technical Help</h4><p>I had several groups of non-technical founders ask about outsourcing technical help. For some people, this might work.</p><p>In my experience, it hasn’t.</p><p>Outsourcing is tricky because you’re ultimately hiring a mercenary. The person might be really talented, but their perception of involvement is fundamentally different from yours.</p><p><strong>They look at it as a job, and you look at it as your product.</strong></p><p>Bringing in technical team members that are invested in the product is essential. Yes, it will cost equity. But it ensures a team that’s willing to do the extra things, and not just those they’re explicitly getting paid for.</p><p>People you know are always the best first candidates. People those folks recommend come next.</p><h4>4: When to go full-time</h4><p>Some founders I spoke with had gotten off the ground, but weren’t working on the startup full-time yet. This is a really tricky situation for some people, because it usually means making the call between a stable, known salary and the great unknown.</p><p>I battled with this decision myself, and empathize with the difficulty. It’s <em>really hard</em> to make decisions that challenge your quality of life.</p><p>Everyone’s answer to this question is different, but generally there’s a few things I’ve found helpful to think about:</p><ol><li>If you’re already working on a startup on the side, there’s something you’re not getting from your current job.</li><li>If you’re already working on a startup on the side, you’re probably spending less time doing your current job well.</li><li>If you’re working on a startup on the side, you’re spending less time giving the startup a chance to succeed.</li></ol><p>There are legitimate reasons for not being able to dive all-in. But most of the time, these are reasons we create for ourselves, and not actual barriers. There’s no one stopping us from resigning — it’s simply uncomfortable.</p><p>It can be a scary, scary thing to go full-time. And having some sort of investment is always helpful in making that easier.</p><p>But at the end of the day, <strong>a side-project startup will have a hard time becoming a full-time startup until you decide to treat it that way.</strong> And for most people, that means committing to it full-time.</p><h4><strong>5: Telling your story</strong></h4><p>This is the last piece of conversation that came up with nearly every early-stage startup I spoke with. In the first five minutes of our talk, I’d ask them what they were building. While everyone <em>had</em> an answer, very few companies were <em>done answering</em> in those first few minutes.</p><p>Being able to get down to an “elevator pitch” of your product is essential in telling your story. It’s the answer you give someone at a bar when they ask you what you work on. It’s the answer you give to the aunt you haven’t seen in 8 years. It’s the answer you give to an investor you meet on a plane.</p><p><strong>In 30 seconds, tell me what you’re working on so that I “get” it.</strong></p><p>Every piece of information that doesn’t help meet that goal should be saved for later. If you can’t make someone think “wow, tell me more” in those first few seconds, they won’t need to know the extra information anyway.</p><p>Getting to that solid, go-to pitch is deceptively difficult, and takes lots of practice. Try it out with anyone who will listen. You’ll get a feel for what sounds *right*, what sounds interesting, and what words you’re comfortable using.</p><p>When it comes time to tell your story, you’ll nail it.</p><h4><strong>Final Thoughts</strong></h4><p>The [<a href="http://tomhauburger.com/">26 Hours</a>] project was awesome. I’m planning to do it again next year, plus-one. I want to say thanks to all the companies who reached out, and to the 26 companies I met with for helping me get better at helping other startups.</p><p>In the meantime, if you’d like to talk more about the project, send me an email, or reach out on [<a href="http://twitter.com/thauburger">Twitter</a>].</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3c4a820c8d4" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>