From: aidotengineer
Heath Black, Managing Director of Product at SignalFire, shares insights on building AI teams based on data from SignalFire’s proprietary AI/ML platform, Beacon [02:29:03]. Beacon tracks over 650 million employees, 80 million companies, and 200 million open-source projects, generating market insights and ranking systems [02:33:04]. The goal is to identify and attract the right talent by focusing on filtering, location, timing, and narrative [03:02:04].
1. Filters: Identifying the Right People
When building AI teams, it’s crucial to apply the right filters in talent searches [03:30:30]. SignalFire data reveals significant trends changing how companies should filter for AI talent:
De-credentialization of AI Talent
Over the past decade, there has been a stark de-credentialization in AI hiring [03:49:10].
- Engineers: AI startups are increasingly hiring engineers without PhDs or prestigious schooling [03:53:35].
- In 2015, 27% of engineer hires were from top schools, and 16% had PhDs [04:03:00].
- In 2023, these numbers dropped to 15% and 7% respectively, a 50% decline [04:09:00].
- Research Scientists: While about 40% of research scientists have advanced degrees, this still makes up less than half of those in such roles [04:29:00].
This shift is attributed to a market focus change from foundational ML research (2015) to real-world application, MLOps, product experience, and understanding user interaction with models [04:51:00].
People Mobility and the “AI V-League”
Historically, AI talent was concentrated in large tech companies like Google, Uber, Meta, and Apple [05:30:00]. This has shifted, with talent now massively concentrating in what SignalFire calls the “AI V-League” – a group of nine companies [05:41:00]. These established tech giants are now actively trying to recruit from the “AI V-League” [06:14:00].
Understanding net employee movement between these companies is essential [06:30:00]. For example, OpenAI has a positive flow of talent from DeepMind, while Cohere experiences a negative trend [06:39:00].
Takeaways for Filtering:
- Work experience now far surpasses education in importance [07:04:00]. Focus on a candidate’s body of work, including open-source contributions or personal projects [07:17:00].
- Assess whether a PhD researcher is truly necessary, or if an experienced engineer will suffice [07:36:00].
- Consider removing or making academic requirements “soft” in job postings to broaden the talent funnel to those with relevant experience [07:48:00].
2. Location: Where to Find Talent
Despite debates about the demise of certain tech hubs, location still significantly matters in building AI teams [08:10:00].
- San Francisco: Makes up about 29% of all startup engineers [08:23:00]. While slightly down from its 2013 peak of 33%, it has been ticking up since 2021 [08:30:00]. For big tech, 50% of engineers still reside in the San Francisco Bay Area [08:49:00].
- AI Specifics: San Francisco leads with 35% of all AI engineers [09:00:00]. Seattle accounts for 22%, and New York for 10% [09:11:00]. These markets are “punching well above their weight” in terms of AI hiring and talent concentration [09:28:00].
- Funding Correlation: San Francisco receives nearly 38% of all early-stage funding into AI startups, compared to 26% of total early-stage funding in the U.S [09:48:00]. This indicates a strong correlation between funding and talent concentration.
Takeaways for Location:
- Data, not social media debates, determines a market’s viability [10:15:00].
- Location still matters significantly, even in a distributed work environment [10:22:00].
- San Francisco, Seattle, and New York are the premier locations for AI talent and funding [10:27:00].
- Continuously monitor location and funding markets to observe talent and capital flows [10:33:00].
3. Timing: When to Engage Talent
Finding the right person is as much about timing as it is about talent [10:46:00]. Timing involves knowing when people are likely to leave their current roles and when they are most inclined to join a company at your stage [11:11:00].
Retention Rates and Poachability
Analyzing the retention rates of “AI V-League” companies can indicate when employees are more receptive to outreach [11:32:00]. For example, Anthropic has a 66% four-year retention rate, while Perplexity AI hovers around 43% [11:39:00]. Understanding these rates creates a “poachability score” [12:08:00].
Generational Differences in Job Mobility
Different generations exhibit distinct behaviors regarding job changes [12:16:00].
- In 2023, nearly 27% of Gen Z left their jobs, more than double the rate of Gen X [12:25:00].
- Within four years of graduating, Gen Z averages 2.2 jobs, compared to Gen X’s 1.1 [12:39:00].
- This is partly due to slower promotion rates and layoffs, but largely attributed to Gen Z’s “penchant for people to take risk and bet on themselves” [12:52:00].
Historical Company Composition
SignalFire’s “historical composition” tool provides snapshots of admired companies’ organizational structures at different points in time [13:42:00]. This includes insights into sales leaders, early engineers, and key product launches [13:54:00]. This data helps identify:
- A candidate’s risk profile – are they willing to join a company at your stage [14:10:00]?
- Their motivations [14:17:00].
- Potential for a “10x hire” – an individual who can significantly elevate the team [14:20:00].
Takeaways for Timing:
- Understand timing for both outreach and impact [14:29:00].
- Track competitor retention rates and individual profiles for changes [14:34:00].
- Study generational patterns of job mobility [14:50:00].
- Know when people join and leave companies to identify 10xers and those likely to join a company at your stage [14:59:00].
4. Narrative: Crafting the Right Story
The narrative you present to potential hires is critical, especially as traditional incentives like salary and equity become less dominant [15:10:00].
Moving Beyond Salary and Equity
Historically, pay and equity were the primary components of a recruiting narrative [16:23:00]. However, reliance on these alone is no longer sustainable:
- Salary: From November 2022 to November 2024, average tech salaries only increased by 1.6% [16:32:00]. AI engineers, specifically, command a 5% salary premium and a 10-20% equity premium over other engineering roles, making them even more expensive [16:50:00]. Therefore, salary as a sole selling point must be abandoned [17:17:00].
- Equity: There has been a precipitous decline in the exercise of vested shares [17:27:00]. In Q2 2024, only 33% of people exercised vested shares, down from 55% a few years prior [17:33:00]. This is driven by concerns over high valuations, the cost of liquid capital, and the rapidly shifting AI market [17:47:00].
Emphasizing Non-Monetary Factors
To attract top talent, companies must highlight non-monetary aspects of their offering:
- A close-knit environment and working with founders [18:15:00].
- Collaborative teams [18:20:00].
- Speed and lack of friction in getting work done [18:23:00].
- A big mission [18:25:00].
- Opportunities for personal and career growth [18:29:00].
- Exploding markets and the chance to solve complex problems [18:31:00].
Understanding your company’s “arc” – its triumphs and future direction – is crucial for crafting a compelling narrative [16:06:00].
Takeaways for Narrative:
- Articulate your company’s triumphs, current standing, and future direction (your “arc”) [16:13:00].
- Move beyond salary and equity as primary selling points, as they are no longer sufficient [17:10:00].
- Focus on the unique benefits and opportunities your company offers beyond compensation [18:11:00].
Conclusion
In a competitive landscape where many companies are “fishing in the same engineering pond,” leveraging recruiting data provides a significant edge [18:48:00]. Just as data is used to build AI systems and models, it should be used to build AI teams [18:59:00]. Your team is your most valuable product [19:08:00].
Summary for Recruiters:
- Filter accordingly based on de-credentialization trends [19:13:00].
- Watch where people move as location still matters [19:16:00].
- Use data to identify the right time for outreach and to understand tenure [19:23:00].
- All these factors will help you craft a better narrative [19:31:00].
Summary for Job Seekers:
- Know where people you admire go, not just companies but the space [19:48:00].
- Watch how long they stay there to understand treatment and potential [19:55:00].
- Know what you want in your career arc [20:03:00].