Breakfast

Day 1 begins! If your day-to-day role is in platform or data engineering, join the respective affinity breakfasts in the breakout rooms (affinity breakfasts for applications and front-end engineering are on Day 2). Otherwise, grab some food and a coffee from the breakfast buffet and feel free to eat in the Grove Ballroom.

Day 1 Kick-Off with Vivek

Vivek Vaidya will kick-off Day 1 of the first inaugural super{summit}!

Micro-stories with Microservices: Experiences Around the Hive

An exciting panel of learnings from using microservices in the field. Hear from super{set} colleagues on all their rousing experiences, good and bad.

HOW SPECTRUMLABS USES TERRAFORM & DBT TO ADMINISTER & TRANSFORM IN SNOWFLAKE.

Learn how we at SpectrumLabs manage and administer SnowflakeDB using Terraform and CICD. As bonus we will also show how we practice CICD transforming and loading data in Snowflake using DBT.

Technology Leadership Panel

Moderated by Vivek, hear from 6 leaders across super{set} companies on their approach to leading technology teams across people, product, and customers. Discover how the approach to leadership evolves across the different stages of a startup’s evolution, and unearth lessons for your own career journey.

Lunch Break

Using gRPC throughout a web application

This session is about using gRPC in a web application end to end. Usually people translate from gRPC to JSON for frontend of their web apps and lose out on benefits of gRPC. In this session we will learn to setup a frontend (Vue.js + typescript but easily applicable to react or angular) communicating with the backend api which is serving gRPC. Preparation - Make sure your favorite IDE (Intellij, vim, emacs, vscode etc) can edit some typescript code.

How many 1's and 0's does it take to make a King?

How does a computer “understand” text? What’s the buzz about Natural Language Processing (NLP)? In this hands-on tutorial, bring your laptop and experience turning text into numbers and then into context. No coding experience required. If you can press “shift+enter”, you’re good to go!

When Inference meets Engineering at the Hive

This session consists of short lightning rounds featuring several companies of the Hive that will address the following two questions: What is the inference problem that the company is solving? What is the interface between data science and engineering? The scope will provide a good opportunity for data scientists from the different companies to learn about the type of problems, solutions, and challenges that other companies are working on and at the same time showcase the engineering processes that power machine learning cycles across the Hive.

Statistical Privacy: No Free Lunch

(Guided discussion) An intense battle over the future of the internet is spurring a complete overhaul of the way that a lot of the major players in Big A*%hole Tech (BAT) collect and use personal information. Certain giants have doubled down on privacy and are instituting changes that allow consumers to keep their personal data locked in an airtight safe if they so choose. But some want to have their cake and eat it, too - in particular, those with business models that rely heavily on the ability to advertise. Public assertions made by half of the Big Four claim to be able to maintain the same targeted-ad relevancy while providing more privacy to the users. But is it possible to deliver the same product with fewer resources? Are privacy enhancing technologies (PETs) like differential privacy and multiparty computation the silver bullets that all the recent publicity makes them out to be?

Fireside Chat with Harvard CS Professor James Mickens

Tom Chavez sits down with Harvard Computer Science professor James Mickens for a free-ranging fireside chat covering software development, CS academia, ethical tech, life-long learning, and if we’re lucky some spicy takes on crypto, social media, and all that’s wrong with tech.

Data Affinity Breakfast

Calling all data scientists, data engineers, and data enthusiasts! Grab your breakfast and a coffee from the foyer buffet and join the data experts across the hive for unstructured mingling.

empty

Your relationship with data: what’s the deal?

You work with personal data but do you think about your relationship with people behind the data? In this session, you will participate in a structured conversation about identifying your role and responsibility in building toward data dignity for the people impacted by what you deploy. You’ll learn more about the opportunities for macro and micro level change, with a focus on how you and your teams can reframe how you do your work to make headway on today’s privacy crisis.

Avoid Observability MeLTdown using Metrics, Logs and Traces

Discuss monitoring and observability (o11y) methods across Habu and Ketch that DevOps and SRE teams can employ in monitoring distributed systems with insights into some of the popular open source and vendor specific tooling for the 3 pillars of o11y - metrics, logs and traces.

Empty

empty

Building Scalable Data Pipelines with Argo Workflows

This session will be an overview of how we use Argo Workflows to build elastically scaling data pipelines at Spectrum Labs. We will demonstrate how we have built a framework that allows us to deploy scalable Spark applications with only a few lines of code. Join us to understand how leveraging Argo's container-native engine allows us to build on our existing Kubernetes investment and simplify our development stack.

Safe Blue-Green deploy with istio and knative

The secret to not losing hair and having a happy lifestyle is to have a safe deploy! Imagine a world where you can easily sleep at night! On this session, we will be presenting and doing a demo on how spectrum leverage istio, circle-ci, knative for deployment

Demystifying De-Identification

Workshop with discussion and demo. The session will begin with an overview of de-identification terminology (de-identification, anonymization, redaction, pseudonymization), how these have been misunderstood, and what to think about when choosing between one of these and other privacy enhancing technologies. The attendees should bring a sample dataset (preferably made up of unstructured text) and a use case in mind. Each attendee will receive an API key to process a data sample and we will discuss the results. Data can be in languages other than English. Please confirm with organizer that the language is supported first.

Onboarding and local development of microservices

Different ways to onboard and help local development for microservices in SturdyAI, Ketch and Habu.

Platform Affinity Breakfast

Do you live and breath infra, security, and deployment? Grab your breakfast and a coffee from the foyer and meet like-minded platform talent from across the hive for some unstructured breakfast mingling.

empty

How to automate your on-call and improve your sleep with Robusta

Don’t get your hair in the butter; automate your on-call response with a free open source tool called Robusta. Don’t be a two-cent man and wake up at 3am to silence pesky alerts. Learn how to deploy Robusta in your clusters and start automating alerts in less than 5 minutes.

No Country for Old Front-end DevOps

Writing webapps is fun. The moment you see "It works" (on your machine) is exhilarating! Let's talk about ways to get it to the world and the complexities involved. Let's automate your breaking code to production.

Empty

empty

The Why and How of Patent Protection for Technology Startups

In this session, the Super{set} outside patent attorneys will discuss the strategic and tactical considerations for when and how to pursue patent coverage. We’ll discuss the tradeoffs between patent and trade secret protection, an overview of the patenting process, and some guidance on when to get started. The format will be a blend of fireside chats with inventors who have gone through the process along with some presentation slides.

API Communication

This session is mainly focused on the different ways Boomjam and Habu communicate with APIs. It will cover how Boomjam migrated from GraphQL to REST and how and why Habu uses React Query. The audience will get a chance to see the pros and cons of the technologies used at both companies.

Front-End Tooling Round Up

JavaScript or TypeScript? Redux or React Query? Webpack, Vite, or Snowpack? Sass, CSS modules, CSS-in-JS, or Tailwind?? Do I even need a library for that??? Ever feel overwhelmed by front-end tooling? You're not alone. But don't worry, we're here to help. In this session, panel members will discuss popular front-end tools in terms of the specific challenges and pain-points they have helped to address at their respective companies, and also invite attendees to share their own problems and solutions. The session will also cover tips on how to think about tooling, how to choose between alternatives, and when to "build versus buy". By the end of the session, attendees will have solid principles for deciding whether or not that shiny new framework everyone is raving about is worthy of adoption.

Gunfight at the Flux Corral

CI (continuous integration) and CD (continuous delivery) are much desired tools yet not very prevalent in the Super{set} Hive. How did Ketch establish and utilize this? What were some of the gunfights along the way?

Breakfast

Day 2 is here! If you do Applications or Front-end Engineering, join your colleagues in the respective breakout rooms for an affinity breakfast. Otherwise, grab some food and a coffee from the buffet and feel free to eat in the Grove Ballroom.

Day 2 Kick-Off with Vivek

Vivek welcomes us to Day 2 of the super{summit}!

Secure Deployment of Infrastructure in Customer Environment -V1

The attendees will learn about how Markov Hybrid Deployment. In the hybrid deployment Markov infrastructure that interacts with customer data is deployed within customer VPC and is OFF by default. We follow security best practices and leverage cloud provided Server-less tech stack for reliability and keeping costs in check for our customers.

Infrastructure Best Practices: Or Else

Learn about Infrastructure best practices when building your startup. From technology recommendations, observability, high availability, cost optimizations, and security.

The privacy stack: a reference architecture to put privacy by design into practice

The Privacy Stack Reference Architecture is an initiative by members of the Ethical Tech Project that outlines key concepts, implementation recommendations, and other best practices for developers and practitioners to design, build, and operate systems and products that place data owners - humans - at the center. In advance of the session you’ll be asked to review a working draft of the reference architecture. The expected time commitment will be about 20-30 minutes. More details to follow. During the session you’ll participate in structured conversations with fellow attendees about your experience working toward implementing privacy by design or building systems that are compliant with privacy regulations.

QUALITY: THE MOST IMPORTANT FEATURE WE CAN BUILD

A BRAINSTORMING SESSION FOR IDEAS, TRICKS AND CAVEATS AROUND TESTING AND SOFTWARE QUALITY.

Client Specific Inferencing with Reusable Logic

There is not really a one-size-fits-all inferencing approach that handles multiple different use cases. There's always some kind of tuning for a particular use case. How do you provide inferencing that's appropriate for a clients specific needs without compromising other clients use cases? What if the inferencing is really similar between the two, but just different enough that they're not compatible? We have run into this use case at Spectrum Labs and have a scalable and performant way of handling these client-specific needs. We have built configurable inferencing logic that data scientists can leverage in a client-specific way without the engineering team needing to do client-specific code. In this section we will discuss the approach used at Spectrum Labs and some lessons learned.

Fireside Chat with Glean CEO Arvind Jain

Vivek sits down with Arvind Jain, currently CEO of Glean and previously founder of Rubrik, Inc and a Distinguished Engineer at Google, to learn about his lessons from 11 years at Google, becoming a founder, and leading an enterprise AI company.

Group Photo

All summit attendees are asked to wear a western style or “cowboy cut” shirt!

Front-end Affinity Breakfast

Calling all front-end engineers and designers! Grab some breakfast chow and mingle with front-end peers from across the hive in this unstructured networking session.

empty

Modern Front-End Tech Stack

The session is mainly a demo session. It walks through the front end tech stack of Habu and Boomjam use, and the reason these technologies, libraries, and tools are picked. It also covers the file structure, testing, deployment of the front end project in Habu and Boomjam. The Audience will have a deeper understanding in front end development and be able to pick the right tech stack for their own project after the session.

Third Party Integrations

Compare and Contrast Integration Use-case for Ketch and Habu - Presentation

Empty

Hiring Engineering Talent and Building a Hiring Brand

This will be an open discussion on the importance of building a recruiting and hiring culture early on in the engineering organization. We will discuss topics like how to source candidate, how to engage with a network and how to leverage resources within super{set} in order to get the most gains out of your efforts. Please come prepared to discuss what has worked or not worked in your hiring efforts.

WHAT WE CAN LEARN FROM DATA BEFORE DOING ANY MATH

THIS SESSION AIMS TO SHOW WHAT WE DO AS PART OF THE PRE-SALES PROCESS AT SPECTRUM, TO QUICKLY FIND AND VISUALIZE TRENDS ACROSS POTENTIAL CLIENTS' DATA, WITH THE OUTPUT GIVING OUR DATA SCIENCE AND ENGINEERING TEAMS AN EARLY HEADS-UP FOR WHAT WILL BE REQUIRED ON THEIR PART AFTER THE CLIENT SIGNS. PARTICULAR ATTENTION IS GIVEN TO FINDING PATTERNS FOR POTENTIAL AUTOMATION, AS MANY OF OUR CLIENTS HAVE THE NEED TO SCALE. THESE FINDINGS ALSO CONTRIBUTE TO FINE-TUNING OUR AI BEHAVIOR SOLUTIONS FOR CURRENT CLIENTS TO MAXIMIZE CLOSING OUR FEEDBACK LOOP BETWEEN DATA ANALYTICS AND DATA SCIENCE.

Empty

Empty

Applications Affinity Breakfast

Microservices chatter needs coffee. Grab your breakfast and join the room for an unstructured networking session on application building!

empty

How is Snowflake used in The Hive

We will start with a quick intro to Snowflake and some of the buzz words around it. Three presenters will discuss how companies in the hive are using Snowflake both in production and in customer visualizations. The format will be 3-4 short presentations. There will be plenty of time for questions you have. You won't learn everything there is to know about these topics, but you'll come away with some ideas and some contacts who could help you delve further into applying these techniques in your situation.

Experimentation for the Modern Advertiser and Marketer

No dev environment setup is needed for this session. This will be a classic lecture of slides followed by an open discussion. This session will go through how Habu is trying to design strong A/B tests for advertisers and marketers to use once Chrome (if ever, as they've stated) removes cookies. The methodology we have found and implemented has no individual targeting; instead is based off of geos, or various cartographical regions. This session will briefly cover the algorithm used, how we proposed to use this experiment for various clients, how it was received, fixes, improvements, and how we want to take it further. Discussions will revolve around how this type of methodology could be leveraged by other companies even if experimentation is not of use and how we at Habu can leverage expertise from other individuals to improve speed and execution. We believe geo-based targeting and experimentation is the future of measuring lift and "quality of advertising" for companies and this session will explain how.

Empty

A Song of UX & Dev

Like ice and fire, UX and development can be on the opposite ends of the spectrum, with conflicting approaches toward obtaining the iron throne of successful product delivery. They can also be harmoniously complimentary in their pursuit of delivery dominance, slaying dragons with the greatest of ease. In this session, we’ll explore common perils that impede design and development, as well as recommendations for avoiding them like Greyscale. We’ll also take a brief peek behind the UX curtain to better understand what the heck happens before development gets involved. Get ready friends, wireframes are coming!

Get your design to code

- Small intro to design system and should you implement it? - Multiple possible ways in which one can go about scaling design system/UI styling throughout the journey of the product. - Go around the table to discuss what has worked well or not for everyone - Figure out the key factors involved to make this decision. - How to figure out what approach is best. Should you build something from the ground up? - - Or should you pick a framework and build upon that? - What are the tradeoffs? - Once you pick up an approach, how smooth is the transition if the need arise? [Optional] Attendees can come prepared with some code snippets which would help make the point clear.

Empty

Empty