Startups

5 machine learning essentials nontechnical leaders need to understand

Comment

Jumble of multicoloured wires untangling into straight lines over a white background. Cape Town, South Africa. Feb 2019.
Image Credits: David Malan (opens in a new window) / Getty Images

Snehal Kundalkar

Contributor

Snehal Kundalkar is the chief technology officer at Valence. She has been leading Silicon Valley firms for the last two decades, including work at Apple and Reddit.

We’re living in a phenomenal moment for machine learning (ML), what Sonali Sambhus, head of developer and ML platform at Square, describes as “the democratization of ML.” It’s become the foundation of business and growth acceleration because of the incredible pace of change and development in this space.

But for engineering and team leaders without an ML background, this can also feel overwhelming and intimidating. I regularly meet smart, successful, highly competent and normally very confident leaders who struggle to navigate a constructive or effective conversation on ML — even though some of them lead teams that engineer it.

I’ve spent more than two decades in the ML space, including work at Apple to build the world’s largest online app and music store. As the senior director of engineering, anti-evil, at Reddit, I used ML to understand and combat the dark side of the web.

For this piece, I interviewed a select group of successful ML leaders including Sambhus; Lior Gavish, co-founder at Monte Carlo; and Yotam Hadass, VP of engineering at Electric.ai, for their insights. I’ve distilled our best practices and must-know components into five practical and easily applicable lessons.

1. ML recruiting strategy

Recruiting for ML comes with several challenges.

The first is that it can be difficult to differentiate machine learning roles from more traditional job profiles (such as data analysts, data engineers and data scientists) because there’s a heavy overlap between descriptions.

Secondly, finding the level of experience required can be challenging. Few people in the industry have substantial experience delivering production-grade ML (for instance, you’ll sometimes notice resumes that specify experience with ML models but then find their models are rule-based engines rather than real ML models).

When it comes to recruiting for ML, hire experts when you can, but also look into how training can help you meet your talent needs. Consider upskilling your current team of software engineers into data/ML engineers or hire promising candidates and provide them with an ML education.

machine learning essentials for leaders
Image Credits: Snehal Kundalkar

The other effective way to overcome these recruiting challenges is to define roles largely around:

  • Product: Look for candidates with a technical curiosity and a strong business/product sense. This framework is often more important than the ability to apply the most sophisticated models.
  • Data: Look for candidates that can help select models, design features, handle data modeling/vectorization and analyze results.
  • Platform/Infrastructure: Look for people who evaluate/integrate/build platforms to significantly accelerate the productivity of data and engineering teams; extract, transform, load (ETLs); warehouse infrastructures; and CI/CD frameworks for ML.

Again, consider the power of training — an engineer with the right curiosity and interest can become the ML expert you need.

Regularly engaging with industry advisors and academics is another way to provide the team with updates on the latest and greatest approaches to ML. Quality bootcamps can be a great way to upskill your teams.

2. Organizational structure

How to best structure the role of the ML team within the larger organization is a significant decision that impacts the efficiency and predictability of the business and should be guided by the stage and size of the company.

Early stage (<25 members): At this size, a shared central team is the safest and quickest way to develop infrastructure and organizational readiness. In the early stage, your ML team should constitute 10%-20% of the entire engineering team.

Midstage (25-500 members): By midstage, it’s best to focus on vertically integrated teams. Gavish is a huge fan of vertical ML teams “because they have a huge advantage in terms of gaining deep understanding of the problem being solved.”

A vertical integration also allows for sustained focus and prioritization, which is needed because midstage ML projects tend to be longer and more uncertain.

Mature (500+ members): At this stage, the business should create a separate ML platform/infra team. For example, Square is a 2,500-plus engineering org with over 100 data scientists and ML engineers and more than 15 ML platform/infra engineers. The ML teams are aligned with individual business units such as chatbots, risk/fraud detection, etc., rather than specific technology. And they have an ML platform/infra team shared across other teams in the company.

However, remember that the size of the team varies depending on how key ML is to the product and services being developed.

3. ML pipeline

Deploying and maintaining ML pipelines is not dramatically different from deploying and maintaining general software. ML knowledge is required around building, tuning, testing, verifying and versioning the model — as well as monitoring it.

The key steps to successfully build, deploy and maintain an ML pipeline are:

  • Define a product problem and determine a fit for ML.
  • Refine datasets.
  • Know and isolate data issues versus model drawbacks.
  • Test, debug and version your models.

Using off-the-shelf software can be an incredibly effective way to reduce the cost and dependency on highly skilled and specialized ML engineers, but be careful of unintentionally creating a disorganized spaghetti solution suite that is difficult to maintain.

While the industry is nascent, tools like Databricks, AWS SageMaker, Tecton, Cortex and others will save time and resources. As far as platforms and libraries, there are many competing solutions in the market: TensorFlow, PyTorch, Keras, Scikit-learn, Pandas, NLTK, etc.

4. Metrics and evaluation

The key challenge around ML is reliability. How can you be sure your model is performing adequately before it’s deployed? How do you monitor production performance and troubleshoot issues? The solution is pretty similar to software engineering: observability.

It’s critical to monitor and track application performance. Hadass recommends “Building Machine Learning Powered Applications” by Emmanuel Ameisen to understand how to do so.

A model that performs better than a baseline (where there is no ML) and is both stable and secure should be good enough to take to production. As a framework, I would vouch for iteration over perfection.

Rolling out models under a feature flag is safe and ensures that you can turn it off quickly before disaster hits. The ability to run multiple versions via A/B testing of the model in production will drastically increase the confidence in the new model and will guarantee an overall higher level of reliability.

A good dataset is a must. It should be one that is meticulously created and reflects production scenarios. Build a system that allows you to backtrack against historical datasets and compare with predictions made by previous versions of the model.

You need metrics and evaluation to address concerns around good models versus bad, such as:

  • Usefulness to end user.
  • Data security.
  • Stability of the model.
  • Practicality of the predictions and recommendations.
  • Ability to explain why a model made the recommendation it did.

5. Common pitfalls

On first read, some of these pitfalls may seem like common sense, but they are worth both reiterating and reflecting on since they can help guide your team to making the best decision during a critical moment.

Don’t:

  • Apply ML to problems that aren’t a good fit for ML, like straightforward sequence of steps.
  • Expect instant results. Impactful ML takes patience and iterations to get solid results.
  • Focus on model success metrics without enough attention to product success metrics.
  • Underestimate the tooling and infrastructure costs leading to slow engineering progress.

Within the last decade, ML has established itself as a technology accelerator. It’s critical in driving automation and bottom-line profitability and growth. This necessitates the need for leaders to know and embrace ML and keep up with the lightning-speed advances in ML technology.

Integrating ML teams effectively into the business starts with an understanding of what makes the right candidate and how to structure the team for maximum velocity and focus.

Leaders should focus on guiding the team to build end-to-end models with integrated observability and monitoring before the models hit production. Evaluate models based on product success, not model success. Avoid common pitfalls under high-stress situations by being intentional about monitoring for them and proactively engaging industry experts and academics to help keep the team up to date on the latest developments.

To solve all the small things, look to everyday Little AI

More TechCrunch

Government officials and AI industry executives agreed on Tuesday to apply elementary safety measures in the fast-moving field and establish an international safety research network. Nearly six months after the…

In Seoul summit, heads of states and companies commit to AI safety

Copilot, Microsoft’s brand of generative AI, will soon be far more deeply integrated into the Windows 11 experience.

Microsoft wants to make Windows an AI operating system, launches Copilot+ PCs

Some startups choose to bootstrap from the beginning while others find themselves forced into self funding by a lack of investor interest or a business model that doesn’t fit traditional…

VCs wanted FarmboxRx to become a meal kit, the company bootstrapped instead

Uber and Lyft drivers in Minnesota will see higher pay thanks to a deal between the state and the country’s two largest ride-hailing companies. The upshot: a new law that…

Uber’s and Lyft’s ride-hailing deal with Minnesota comes at a cost

Andreessen Horowitz’s American Dynamism fund has established a new fellowship program aimed at introducing top engineers and technologists to venture investing, a move that could help the firm identify less…

a16z’s American Dynamism team launches program to introduce technical minds to VC

Another fintech startup, and its customers, has been gravely impacted by the implosion of banking-as-a-service startup Synapse. Copper Banking, a digital banking service aimed at teens, notified its customers on…

Teen fintech Copper had to abruptly discontinue its banking, debit products

Autodesk — the 3D tools behemoth — has acquired Wonder Dynamics, a startup that lets creators quickly and easily make complex characters and visual effects using AI-powered image analysis. The…

Autodesk acquires AI-powered VFX startup Wonder Dynamics

Farcaster, a blockchain-based social protocol founded by two Coinbase alumni, announced on Tuesday that it closed a $150 million fundraise. Led by Paradigm, the platform also raised money from a16z…

Farcaster, a crypto-based social network, raised $150M with just 80K daily users

Microsoft announced on Tuesday during its annual Build conference that it’s bringing “Windows Volumetric Apps” to Meta Quest headsets. The partnership will allow Microsoft to bring Windows 365 and local…

Microsoft’s new ‘Volumetric Apps’ for Quest headsets extend Windows apps into the 3D space

The spam reached Bluesky by first crossing over two other decentralized networks: Mastodon and Nostr.

The ‘vote Trump’ spam that hit Bluesky in May came from decentralized rival Nostr

Welcome to TechCrunch Fintech! This week, we’re looking at the continued fallout from Synapse’s bankruptcy, how Layer wants to disrupt SMB accounting, and much more! To get a roundup of…

There’s a real appetite for a fintech alternative to QuickBooks

The company is hoping to produce electricity at $13 per megawatt hour, which would be more than 50% cheaper than traditional onshore wind.

Bill Gates-backed wind startup AirLoom is raising $12M, filings reveal

Generative AI makes stuff up. It can be biased. Sometimes it spits out toxic text. So can it be “safe”? Rick Caccia, the CEO of WitnessAI, believes it can. “Securing…

WitnessAI is building guardrails for generative AI models

It’s not often that you hear about a seed round above $10 million. H, a startup based in Paris and previously known as Holistic AI, has announced a $220 million…

French AI startup H raises $220M seed round

Hey there, Series A to B startups with $35 million or less in funding — we’ve got an exciting opportunity that’s tailor-made for your growth journey! If you’re looking to…

Boost your startup’s growth with a ScaleUp package at TC Disrupt 2024

TikTok is pulling out all the stops to prevent its impending ban in the United States. Aside from initiating legal action against the U.S. government, that means shaping up its…

As a US ban looms, TikTok announces a $1M program for socially driven creators

Microsoft wants to put its Copilot everywhere. It’s only a matter of time before Microsoft renames its annual Build developer conference to Microsoft Copilot. Hopefully, some of those upcoming events…

Microsoft’s Power Automate no-code platform adds AI flows

Build is Microsoft’s largest developer conference and of course, it’s all about AI this year. So it’s no surprise that GitHub’s Copilot, GitHub’s “AI pair programming tool,” is taking center…

GitHub Copilot gets extensions

Microsoft wants to make its brand of generative AI more useful for teams — specifically teams across corporations and large enterprise organizations. This morning at its annual Build dev conference,…

Microsoft intros a Copilot for teams

Microsoft’s big focus at this year’s Build conference is generative AI. And to that end, the tech giant announced a series of updates to its platforms for building generative AI-powered…

Microsoft upgrades its AI app-building platforms

The U.K.’s data protection watchdog has closed an almost year-long investigation of Snap’s AI chatbot, My AI — saying it’s satisfied the social media firm has addressed concerns about risks…

UK data protection watchdog ends privacy probe of Snap’s GenAI chatbot, but warns industry

U.S. cell carrier Patriot Mobile experienced a data breach that included subscribers’ personal information, including full names, email addresses, home ZIP codes and account PINs, TechCrunch has learned. Patriot Mobile,…

Conservative cell carrier Patriot Mobile hit by data breach

It’s been three years since Spotify acquired live audio startup Betty Labs, and yet the music streaming service isn’t leveraging the technology to its fullest potential — at least not…

Spotify’s ‘Listening Party’ feature falls short of expectations

Alchemist Accelerator has a new pile of AI-forward companies demoing their wares today, if you care to watch, and the program itself is making some international moves into Tokyo and…

Alchemist’s latest batch puts AI to work as accelerator expands to Tokyo, Doha

“Late Pledge” allows campaign creators to continue collecting money even after the campaign has closed.

Kickstarter now lets you pledge after a campaign closes

Stack AI’s co-founders, Antoni Rosinol and Bernardo Aceituno, were PhD students at MIT wrapping up their degrees in 2022 just as large language models were becoming more mainstream. ChatGPT would…

Stack AI wants to make it easier to build AI-fueled workflows

Pinecone, the vector database startup founded by Edo Liberty, the former head of Amazon’s AI Labs, has long been at the forefront of helping businesses augment large language models (LLMs)…

Pinecone launches its serverless vector database out of preview

Young geothermal energy wells can be like budding prodigies, each brimming with potential to outshine their peers. But like people, most decline with age. In California, for example, the amount…

Special mud helps XGS Energy get more power out of geothermal wells

Featured Article

Sonos finally made some headphones

The market play is clear from the outset: The $449 headphones are firmly targeted at an audience that would otherwise be purchasing the Bose QC Ultra or Apple AirPods Max.

11 hours ago
Sonos finally made some headphones

Adobe says the feature is up to the task, regardless of how complex of a background the object is set against.

Adobe brings Firefly AI-powered Generative Remove to Lightroom