Welcome to MGMT Boston where we try to help (500+) of you manage your awareness of top Boston startups and local up & coming operators putting in the work. Glad to have you here!
TLDR:
Neural Magic - helps organizations deliver AI through software rather than expensive and inefficient hardware like GPUs
Thanks to Sarah H. for the intro to Neural Magic
Max Milhan, Lead Product Manager @ Rhino - a PM building more intentional roadmaps by sharing wins and progress, tightening the screws each week, to chart a better course
Thanks to Brian S. for the intro to Max
Other Resources:
The Endeca Effect: Overview / Markets / People / Products / Conclusion / Bonus - Steve Papa Alumni Learnings
MGMT Boston Operators Club - e-mail me directly to learn more!
Q3 Startups Highlighted: Snyk, FeatureByte
Q3 Operators Highlighted: Alexa Murray / WHOOP, Kieran O’Driscoll / AtScale
Q1 Startups & Operators Highlighted / Q2 Startups & Operators Highlighted
Collecting some feedback around how you’re enjoying the newsletter so far. If you’ve got a minute or two to spare, please check out this Merlin powered survey!
Neural Magic
Founders: Alex Matveev & Nir Shavit
Founding: 2018
Mission: Unlock the full potential of your ML environment. Accommodate the continuous growth of neural networks without added complexity or cost
Employees: 53 & ~50% Local
Workplace: Hybrid
Stage & Capital Raised: Series A & $50M raised
Investors: Andreessen Horowitz, Amdocs, Comcast Ventures, NEA, Pillar VC, Ridgeline Ventures, Verizon Ventures, VMware
Key Partners and Customers: Striveworks, Intel, AMD, DigitalOcean, Google Cloud, AWS
Glassdoor Rating: N/A
Valuation (estimated): $100M – $300M (assuming they sold ~20% of the company in the $30M Q4 ‘21 Series A fundraise)
^ this is a useless number. There is no tangible valuation until the business is sold or goes public. Don’t forget it!
Neural Magic helps organizations deliver AI through software rather than expensive and inefficient hardware like GPUs. MIT Professor Nir Shavit and his PHD student Alex Matveev were doing deep learning research at an MIT Lab in 2016 but didn’t have access to the same high powered hardware resources their corporate counterparts had to do traditional neural network computer science research.
First, let’s set the stage. A GPU (graphics processing unit) is a specialized chip that allows you to execute large neural networks to predict desired outcomes. Like a race car engine, more or less. The NVIDIA A100, the type of industrial level chip used to train a LLM (large language model) like OpenAI’s ChatGPT, retails for $10-$15k. For one chip! Some analysts project OpenAI would have needed a cluster of 10,000 GPUs (chips) to train the initial ChatGPT model. I’ll let you do the math.
Quibble on the specifics all you wish, but the point is that it’s unreasonable to think infrastructure costs can continue to scale linearly (or exponentially). For any company, even one with the vast resources of OpenAI & Microsoft, rising infrastructure costs are a problem.
Without the budget to afford GPUs, Nir & Alex built software that optimizes deep learning models to run on CPUs (commodity-level hardware). Scarcity breeds creativity, after all. Spinning out of the university in 2018, the duo founded Neural Magic back when AI was still called machine learning to help AI teams run better, faster deep learning models on a budget.
There are some other secular trends that act as tailwinds for Neural Magic too. Moore’s Law is coming to an end (we think). For 50+ years the advancements in chipmaking have made software development easier on the back of increasingly more powerful hardware. But it’s getting more expensive and more difficult to make smaller and smaller transistors. Our power grid & infrastructure is being pushed to its limits too. Technologists need to come up with new, novel ways to get more efficient. We need better tools to help optimize AI models and software.
Neural Magic is building one of those solutions. Nir, Alex & team are building a bridge between expensive hardware and commodity servers you can rent on cloud providers like Azure, AWS, or Google Cloud. They help organizations optimize their AI use cases to be deployed at very fast speeds by leveraging a well-researched ML optimization practice known as sparsity. Sparsity allows developers to reduce the computational requirements of ML models. Magic? No, computer science. Once model computational needs are reduced, Neural Magic’s inference runtime software, called DeepSparse, executes the model using sparsity-aware techniques and only CPUs, at GPU speeds.
Brian Stevens, the former CTO of Red Hat and Google Cloud, was brought in as Neural Magic’s CEO in 2021 to help lead their commercialization efforts with an emphasis on adopting an open source first model and building out their open source community efforts. Today, any developer can access SparseZoo, an open source repository of existing models that have been developed using Neural Magic’s toolkit. Next, their SparseML tool allows open source developers to optimize custom models and tweak the machine learning to apply to individualized use cases with both private and public data.
Last, developers can run their models in DeepSparse, Neural Magic’s inference runtime. DeepSparse achieves its performance using breakthrough algorithms that reduce the computation needed for neural network execution and accelerate the resulting memory-bound computation, cutting up to 90% of a model’s compute requirements to get more flexibility, faster by optimizing once and then deploying at scale.
In 2023, the team is further building out their revenue engine and adding to an open source community of 1,400+ Slack members. In 2023 YTD they’ve seen over 1M SparseZoo model downloads, more than 2x what they saw in 2022.
Operators to Know (Locally):
Jeannie Finks, Head of Customer Success
Rob Fitzgibbon, UX Lead
Benjamin Fineran, ML Engineering Manager
Michael Goin, Product Engineering Lead
Danny Guinther, Full Stack Tech Lead and Manager
Dan Huang, Software Engineering Manager
Nicole Kim, Head of Product Marketing
Mark Kurtz, Director of Machine Learning
Kierstin Mallett, Business Development Lead
Bill Nell, Software Engineering Manager
Tuan Nguyen, Senior ML Research Engineer
Tyler Smith, Software Engineering Team Lead
Saša Zelenović, Head of Marketing
My investigative powers continue to need work so apologies to the Neural Magic team if I missed many up & coming operators internally
Key Roles To Be Hired:
If I were interviewing here are some questions I’d ask:
What are the key defensible technical advantages that Neural Magic is building?
Who are the key competitors Neural Magic faces in the years ahead?
What are the key milestones for 2023 and what is the long term vision for the company?
What are the biggest priorities as you scale the team? What are the most important roles you’ll be looking to add in 2023?
We’re optimizing for readability here so to learn more about Neural Magic you’ll have to D.Y.O.R. I’m excited to watch this team help more AI companies get more efficient with their infrastructure costs. All consumer and high powered machine learning enthusiasts applaud your efforts. See you around town!
Max Milhan, Lead Product Manager @ Rhino
Career Summary / Sharing Progress = Finding Solutions / Career Insights
Max Milhan is building more intentional roadmaps by sharing wins and progress, tightening the screws each week, to chart a better course. As the Lead Product Manager at remote fintech company Rhino, Max is on the front lines of helping property managers and renters alike build a better renting experience.
Max was a Finance major at the University of Colorado, where he assumed that your first job out of school would be adding the word “analyst” to whatever your major was.. Of his post graduate job offers, technology consulting and the array of experiences it offered seemed the most interesting.
He joined Perficient and became a certified scrum master. Embedded within a telecom company, Max helped them build software in a similar capacity to what he later learned was a Product Manager. Through his various projects, he helped teach software engineering companies how to adopt scrum (an agile software building methodology).
Next, he moved over to management consulting at RAS & Associates. After a few years he reflected back on his work and realized he had the most fun helping build software.. When a friend mentioned that his startup ParkiFi was looking for someone with “agile” experience, Max jumped at the opportunity.
ParkiFi helped track, analyze, and fill open parking spaces using a combination of hardware and software tools. This was Max’s first taste of early stage startups and there were no billable hours in sight. The team built and embedded parking sensors into spaces that could detect a car above them, transmitted that data to a gateway with a cellular modem, which sent data to the cloud to provide parking lot operators a real-time view of their parking lots.
He managed an array of operational responsibilities and got plenty of up close product experience too. But ParkiFi eventually shut down and Max moved north to Boston.
Through a Boston Product Management group event at Drift, Max heard how PMs described their roles and it helped flip a switch in him to tell a better story about how to better position his experience. That community was hugely helpful to his Boston onboarding and he credits Adam Sigel for running Boston Product and Jake Cohen for getting him into the Slack group in the first place.
In Boston, he joined another early stage startup called Authess, building machine learning tools in the edtech space. It was his first true Product Manager experience where he got to do a bit of everything - product management, design, product delivery, QA & client services.
He lived in Brighton and rode the T to and from work for a couple years before moving to East Boston. He loved frequenting Cunard Tavern and Santarpio’s in the neighborhood if you ever find yourself over that way.
Next, Max moved to Dispatch, a software company for large enterprises that managed networks of service technicians. His first week on the job he nervously pulled his boss aside to let him know he thought there might have been a mistake. Dispatch’s customer support team was reading json logs and the account managers were writing SQL queries. Max couldn’t do either!
So every Friday his boss, Sam Robinson, would sit down with him next to a couple of drinks and dig into the code. Max learned to ship code into production and now, when a feature works great but the button placement is wrong, Max can just jump in and fix it himself.
Fintech & real estate were verticals that always appeared to Max because, on the side, he’s a small real estate investor. The thought of combining software and financial products for property managers and renters made a lot of sense. In lieu of placing a security deposit down, Rhino helps make apartment moves and requirements like ahem, broker fees, ahem easier to handle financially. So instead of putting down a whole security deposit up front, you just pay a monthly fee for a financial product.
Max joined the company through a period of rapid growth and helped the team get more crisp on their north star metrics across their engineering pods. Starting from a vague goal of “making property managers happy & using the product more” to moving the activation rate from “x% > y%”, he’s helped forge a strong partnership with analytics to measure everything they do. High performing product development is driving impact you can measure.
Max is responsible for the activation of business in their B2B(2C) model. He helps make it more turnkey for property managers to offer their financial products to more renters. He also helped Rhino identify a company that had a solution to handle traditional deposits they ended up acquiring in January of 2023. Now, they have one place to satisfy a security deposit whether customers choose the insurance product or the traditional deposit / escrow product.
Sharing Progress = Finding Solutions
Max makes an intentional effort each week to share his personal and team’s progress in order to help source solutions to the problems his team is tackling. He has found that sharing updates is a critical exercise in a product management organization and even more important in a remote culture.
He shares what he and the engineers are working on publicly so peers across the company can provide input and course correct as needed.
He makes sure to share wins, misses, and any roadmap delays so that anyone else across the company can relay potential issues or information to help prioritize tradeoffs. Sharing wins builds momentum across teams. It makes goals feel more achievable and results in more wins. Sharing issues early allows you to make trade-offs in plans.
If it feels salesy to talk about accomplishments, Max shares that the best way to flip that script is to think “I want to highlight all the good work the engineers and designers are doing”.
As the week wraps up, Max reports out on the progress they’ve made, changes to their plans, and what actually happened. It only takes 15 minutes to send out a weekly update on their various pod Slack channels to keep visibility high and the messaging consistent.
Career Insights / Learnings
Decision Making Framework - “Inspired by Jeff Bezos’ Regret Minimization Framework I recalled when I thought about joining a startup for the first time, the offer was a little bit less than I was making and I didn’t know how to value the equity component. I made a pros and cons list of how to make it work. I asked friends for advice. I was so stuck on what to do. Finally, one friend stopped me and said ‘does this get you closer or farther to the work you want to be doing in 5 years? If it’s closer, you’ll make it work.’ That advice made my decision so much more clear. A lot of decisions that we think are analytical are actually emotional decisions”
Hold Your Peers Accountable - “It’s ok to protect your team. But you need to take responsibility and hold your team accountable for results too. If you are doing someone else’s work, it will keep you from scaling your impact. It’s ok to hold your peers accountable. Not in a mean way but in a radical candor kind of way. It’s hard for conflict avoidant people but very necessary in a high performing team”
Notch Your Wins - “Keep a victory notes files, whether it’s client feedback or a good review, and keep them in a file somewhere in the cloud, not on your work machine. Any time you have success, put it in there. It makes it easy to fill out performance reviews and is a great place to refer back to when you’re feeling imposter syndrome.”
Being Your Fun Self - “Usually at a new job it takes me a couple months to let down my professional facade. But, one day, inevitably a weird quirk of mine will show itself and people are overwhelmingly relieved to know that I’m another weird human trying to figure it out at work. If we all put on professional facade, we’re missing out on genuine connections at work”
What does Max want to achieve in his career? Happiness. One day it might look like leading a large product team at a growing software unicorn. Or it could be owning a little software company in an important niche that employs 10 people. Either way, Max is continuing to tackle hard problems that need creative solutions.
If you want to learn more about Max, you can find him on LinkedIn. Thanks for sharing. Excited to see all the great work you continue to produce and products you help shape in the quarters and years ahead, notching wins and sharing updates along the way!
Any feedback for me? One thing you liked? One thing you didn’t? Local startups or operators to highlight? Just reply to this e-mail!
See you next week!
-Matt