Where the AI Jobs Are: 6 Companies Shaping Hiring Through 2028

Anthropic entered 2026 with 452 open roles and told investors it could hit $70 billion in revenue by 2028. Google DeepMind grew from 2,500 to 6,000 employees in two years. Sierra, a three-year-old startup, signed San Francisco's largest office lease since OpenAI's 2024 expansion after hitting $150 million in ARR. Meanwhile, Meta cut 8,000 workers and simultaneously restructured 1,000 employees into new roles with titles like "AI builder" and "AI pod lead." These six companies are not hiring the same way, for the same roles, or with the same urgency. Here is what the data shows about where AI talent is going through 2028.

Six Companies, Six Completely Different Hiring Strategies

The AI hiring boom is real, but framing it as a single wave misses the most important detail. There are distinct hiring profiles emerging across frontier labs, applied AI startups, and enterprise AI divisions. Each of the six companies covered here represents a different bet on what kind of talent builds durable AI advantage, and understanding those differences is the single most useful piece of information a senior professional can have before deciding where to focus their search.

The separation between tier one and tier two is sharper than most people realize. Google DeepMind, xAI, and Meta are operating at infrastructure and research scale measured in thousands of employees and hundreds of billions of dollars in capital expenditure. Anthropic is competing at the frontier with a fraction of that headcount and winning on revenue efficiency. Sierra and Decagon are growing faster than any of them on a percentage basis, operating with lean teams and genuine enterprise traction. All six are hiring aggressively. None of them want the same person.

Anthropic: Selective by Design, Expanding by Revenue

Anthropic's headcount surged from roughly 500 employees in late 2023 to approximately 2,300 by late 2025, according to Fortune. What makes that number remarkable is the revenue it sits alongside. Anthropic is on track to hit nearly $10 billion in annualized revenue by end of 2025 and has told investors that figure could reach $70 billion by 2028. No technology company has reached $30 billion in revenue with a headcount this small. Google crossed the same threshold with 32,000 people.

The constraint is intentional. Dario Amodei has been explicit that Anthropic picked coding and enterprise co-work and went deep instead of wide. The hiring profile reflects that: research scientists, safety engineers, ML engineers focused on production systems, enterprise account executives, and trust and safety specialists. Total compensation for senior engineers runs $300,000 to $490,000. Glassdoor rating sits at 4.4, with a 4.8 on compensation and a 3.7 on work-life balance, which tells you exactly what kind of environment it is.

Anthropic recently announced plans to add 200 new jobs by 2027 in its Dublin operation alone, covering engineering, sales, finance, legal, and compliance. It is also currently capturing more than 73% of all spending among companies buying AI tools for the first time. That market position drives hiring demand more than any stated headcount target. If you have a research, safety, or enterprise technical background and want to be at the frontier lab with the clearest path to profitability, Anthropic is the most durable bet in this cohort through 2028.

Google DeepMind: The Lab That Doubled Twice

DeepMind's workforce has swelled to around 6,000 employees, up from roughly 2,500 two years ago, according to the Wall Street Journal. Revelio Labs data puts the year-over-year growth rate at 34.1% from 2024 to 2025 alone, with active job postings up 36.3% over the same period. Sundar Pichai folded Google Brain into DeepMind in 2023, creating the largest concentrated AI research workforce of any company on this list.

The talent war around DeepMind has been intense. Microsoft hired roughly two dozen DeepMind employees in 2025. Meta was reportedly offering $100 million signing bonuses to pull researchers. Google responded by bringing back Noam Shazeer, a foundational figure in transformer architecture who had left in 2021, as part of a licensing deal through Character.AI. Twenty percent of Google AI software engineers hired in 2025 were former Google employees returning, which signals both the strength of the brand and the intensity of external poaching.

DeepMind's current hiring focus spans Gemini model development, robotics and embodied AI, multi-modal systems, and AI safety research. The lab's Genie 3 world model and Gemini Robotics programs are opening new research tracks that did not exist two years ago. For professionals with production ML engineering backgrounds, experience in robotics or computer vision, or deep research credentials, DeepMind represents one of the best combinations of institutional credibility and frontier technical work available anywhere.

xAI: Infrastructure-First, Talent Strategy in Flux

xAI closed a $20 billion Series E round in January 2026 at a $230 billion valuation. Its Colossus supercomputer in Memphis houses 200,000+ GPUs, with plans to scale toward one million. The Memphis workforce alone reached nearly 3,000 employees, covering the full stack from electricians to AI engineers. LinkedIn reports xAI now has just over 5,000 total employees, compared to more than 7,500 at OpenAI and more than 4,700 at Anthropic. In February 2026, xAI was acquired by SpaceX in a consolidation move ahead of a planned IPO.

The hiring picture at xAI is more complicated than the infrastructure numbers suggest. In March 2026, nine of the original eleven co-founders had departed. Elon Musk publicly acknowledged the company "was not built right the first time" and is being rebuilt from the foundations up. SpaceX and Tesla executives parachuted in to evaluate and restructure teams. On the other side of the ledger, two senior engineers from Cursor joined to lead product engineering, and xAI opened a Seattle office with initial roles paying $180,000 to $440,000. The company also secured a $200 million Department of Defense contract in July 2025.

For candidates, xAI is the highest-variance option in this group. The infrastructure scale is genuine and the compute resources are extraordinary. The organizational instability is also genuine. Professionals who thrive in rebuilding environments, particularly those with infrastructure engineering, systems-level ML, or AI product backgrounds, will find real opportunity. Everyone else should watch the 2026 IPO process before committing.

Meta: The Largest AI Headcount Swap in Corporate History

Meta generated $201 billion in revenue in 2025, up 22% year-over-year. Its Q1 2026 revenue came in at $56.3 billion, growing 33% year-over-year. Free cash flow for 2025 was $43.6 billion. And on May 20, 2026, the company began laying off roughly 8,000 employees, approximately 10% of its 78,865-person global workforce, while simultaneously closing 6,000 open roles. This is not a company cutting out of financial distress. It is the most explicit AI-for-headcount swap any major employer has put on paper.

Meta's 2026 capital expenditure guidance tops out at $145 billion, nearly double the $72.2 billion it spent in 2025. Almost all of it is going to AI infrastructure: data centers, compute capacity, and custom silicon. Chief People Officer Janelle Gale's internal memo made the trade explicit: the cuts exist "to run the company more efficiently and to allow us to offset the other investments we're making." About 1,000 employees have already been moved into new roles with titles including "AI builder," "AI pod lead," and "AI org lead." The cuts paid for the pods.

The division that is not being cut is Meta Superintelligence Labs, now run by Alexandr Wang, the 28-year-old founder of Scale AI whom Meta acquired a 49% stake in for $14.3 billion in June 2025. Wang oversees roughly 3,400 people working on frontier model development, including dozens of researchers Meta recruited from OpenAI, reportedly with $100 million signing bonuses. If you are a frontier model researcher, an ML infrastructure engineer, or an AI product leader with large-scale platform experience, Meta Superintelligence Labs is hiring with almost no budget constraint.

Sources: WSJ, Fortune, TechCrunch, Tracxn, Revelio Labs, 2025-2026. Chart by Archer Careers.

Sierra and Decagon: The Startup Tier Nobody Is Talking About Enough

While the attention flows to frontier labs, two applied AI startups have quietly built the fastest-growing hiring pipelines in the industry on a percentage basis. Sierra and Decagon are both attacking the same category: AI agents for enterprise customer experience. They are growing at extraordinary rates from the same starting point, and both are in aggressive expansion mode.

Sierra, co-founded by former Salesforce co-CEO Bret Taylor and ex-Google executive Clay Bavor, hit $150 million in ARR by January 2026, up from $26 million at the end of 2024. That is a nearly 6x jump in twelve months. The company raised a $350 million Series B in September 2025 at a $10 billion valuation, then announced plans to double its headcount and signed San Francisco's largest office lease since OpenAI expanded in 2024. Current headcount sits at 628 (Tracxn, March 2026). Fast Company ranked Sierra the number one Most Innovative company in Applied AI for 2026. Clients include SoFi, Wayfair, Ramp, and Rocket Mortgage.

Decagon was founded in August 2023 by Jesse Zhang and Ashwin Sreenivas. It became a unicorn in 2025, just two years after founding, and raised a $250 million Series D in January 2026 that tripled its valuation to $4.5 billion. As of March 2026, it employs 343 people. Its AI agents handle end-to-end customer support for enterprises including Avis Budget Group, Deutsche Telekom, Notion, Duolingo, and Rippling. One client reduced their support team by 80% after implementing Decagon's platform. The company completed its first employee tender offer at the $4.5 billion valuation, allowing 300+ employees to sell vested shares, a meaningful retention and recruiting signal.

Sources: Sacra, TechBuzz.ai, Bloomberg, 2024-2026. Chart by Archer Careers.

What makes both companies particularly interesting from a career perspective is the role profile they need. Sierra and Decagon are not hiring research scientists or PhDs. They are hiring engineers who can build and deploy reliable production AI systems, forward-deployed engineers and solutions engineers who work directly with enterprise clients, and product managers who understand agentic AI workflows. These are some of the most accessible AI roles for professionals transitioning from enterprise SaaS, customer success, solutions engineering, or B2B product backgrounds.

What This Means for Your Move Through 2028

The six companies above represent three distinct hiring environments, and the one that fits you depends entirely on what you are optimizing for.

If you want frontier research and maximum technical credibility, Anthropic and Google DeepMind are the right targets. Anthropic's revenue trajectory makes it the most structurally sound frontier lab for a multi-year career bet. DeepMind's scale and Alphabet backing make it the most resourced. Both require deep technical credentials: research experience, production ML systems, or safety-specific expertise. Compensation at both starts at $300,000 and climbs quickly. The competitive pressure to win talent away from these two has never been higher, which means negotiating leverage is real.

If you want infrastructure scale and are comfortable with organizational volatility, xAI offers extraordinary compute access and a genuine rebuilding mandate. The IPO timeline through SpaceX creates an equity catalyst that could be significant. The organizational churn is real and documented. This is the right option for engineers who want to build at scale and are not relying on institutional stability.

If you want enterprise traction and startup upside, Sierra and Decagon are the most underrated options in the market. Sierra is doubling headcount from a base of 600 with $150 million in contracted ARR and a $10 billion valuation. Decagon just completed a tender offer at $4.5 billion, which means early employees have already realized liquidity. Both companies have genuine product-market fit, not just investor hype. The roles they need, forward-deployed engineers, AI product managers, solutions architects, and customer success leaders with technical depth, map almost exactly onto the backgrounds of senior professionals at enterprise SaaS companies looking to make a move into AI.

If you want the largest platform and the best compensation floor, Meta Superintelligence Labs is the answer. The $100 million signing bonus headlines are extreme, but they reflect the underlying reality: Wang's team is hiring with essentially no budget ceiling for the right researchers and engineers. The trade-off is organizational complexity and the noise that comes with being inside a 78,000-person company going through a major restructuring. For the right candidate, the equity upside at a $1.3 trillion market cap company combined with the technical resources of Meta's infrastructure is genuinely competitive with any frontier lab.

The professionals who will be best positioned across all six of these companies through 2028 are the ones who can do two things simultaneously: demonstrate hands-on production AI work through a real portfolio, and articulate precisely which company's specific technical bets their background prepares them to advance. The first is a skills problem. The second is a positioning problem. Both are solvable. The window is wide open.

Ready to make your next move?

Archer Careers helps professionals land roles at high-growth startups and top tech companies. From resume and LinkedIn optimization to precision sourcing and offer negotiation, we handle the entire job search so you can focus on what matters.

Book a free 30-minute strategy call at hirearcher.com