Dave Brown to Succeed Meta Infrastructure Chief in 2027 | Magica
Dave Brown to Take Over Meta Infrastructure in Early 2027
Editorial Team
••📖6 min read
Former AWS computing chief Dave Brown will succeed Santosh Janardhan in early 2027 as Meta weighs its next capacity budget after a quarter of negative free cash flow.
Dave Brown, formerly head of AWS computing and machine learning services, will succeed Santosh Janardhan in early 2027.
Meta forecasts $135 billion–$145 billion in company-wide capital spending for 2026; its 2027 capacity plans remain under development.
Reports agree on the infrastructure succession but conflict over who will hold the Meta Compute technical advisory role.
Meta executive Dave Brown, the former senior vice president of computing and machine learning services at Amazon Web Services, will take over the social media company's infrastructure operations in early 2027. He succeeds Santosh Janardhan, Meta's infrastructure chief and co-head of engineering, who joined Facebook in 2009 and helped build its data centers, networks and custom chips.
The handover puts a veteran of Amazon's cloud business in charge of infrastructure serving billions of people as Meta expands computing capacity for AI. It comes between a costly 2026 buildout and decisions about next year's budget, which Meta said in October it was still developing.
Amazon employees in a data center. Brown previously led AWS computing and machine learning services. Photo: Amazon. Source: Amazon.
A cloud operator takes over an established organization
Janardhan announced his decision in a November 11 staff memo. He gave no reason for leaving, according to a report based on the memo. A separate account citing CNBC's reporting said Brown joined Meta in July and Janardhan would remain through the first quarter of 2027 to assist the transition.
Brown spent nearly 19 years at AWS, Amazon's cloud computing business. In Amazon's July departure announcement, Brown described a career that began as a developer and grew into building services used by millions of customers. His former responsibilities covered both computing and machine learning services, experience relevant to Meta's expanding AI operations.
Palantir’s co-founder and CEO says he could back a Democrat in 2028, while excluding a Democratic Socialists of America member. His Democratic history complicates talk of a political conversion; the more immediate question is how his government software company will answer congressional scrutiny.
Janardhan's memo endorsed Brown's experience operating infrastructure for demanding workloads and his judgment about the next set of priorities, according to the account citing CNBC. The succession also builds on Janardhan's existing work: the account drawing on Bloomberg's memo reporting credits him with helping develop the systems now serving billions of people. Neither account establishes that Brown's appointment signals a change in spending strategy.
Company-reported operating and free cash flow. Meta subtracts property and equipment purchases and finance-lease principal payments when calculating free cash flow. Figures cover the whole company. Source: Meta Investor Relations.
Capital spending outpaces quarterly operating cash flow
Meta, whose apps include Facebook, Instagram, Messenger and WhatsApp, said in its third-quarter results that average daily use across its family of apps reached 3.64 billion people in September 2026. Its infrastructure supports that existing business as well as new AI work.
The results show the financial pressure and its limits. The following figures cover the three months ended September 30 and use Meta's definition of capital expenditures, including principal payments on finance leases.
Company-wide measure
Q3 2025
Q3 2026
Capital expenditures, including finance-lease principal
$19.37 billion
$34.91 billion
Operating cash flow
$30.00 billion
$32.79 billion
Free cash flow
$10.63 billion
−$2.12 billion
Capital spending rose about 80% from the same quarter a year earlier, while operating cash flow rose about 9%, calculated from the company's reported figures. Meta calculates its non-GAAP free cash flow by subtracting purchases of property and equipment and finance-lease principal payments from operating cash flow. The negative quarterly figure does not describe the whole year: free cash flow for the first nine months remained positive at $11.05 billion, down from $29.51 billion a year earlier.
Meta forecast full-year 2026 capital expenditures of $135 billion–$145 billion, narrowing the previous $130 billion–$145 billion range. These are company-wide amounts, not a disclosed AI-only budget or an allocation controlled solely by Brown. Meta also reported $82.45 billion in cash, cash equivalents and marketable securities at September 30. It had raised $24.91 billion in net proceeds from issuing long-term debt during the first nine months of 2026.
Credit concerns nevertheless form part of the backdrop. The account citing CNBC said S&P Global Ratings moved the outlook on Meta's AA- credit rating from stable to negative in late October, citing free cash flow among other factors. That was an outlook change, rather than a reported cut to the rating itself.
Meta is also preparing another financing option. In its results, it said its board had approved a new class of non-voting shares, expected to take effect in January 2027, for uses including raising capital, employee compensation and acquisitions. The company linked that flexibility to the scale of investment needed for infrastructure and other AI initiatives, while retaining its founder-led structure. The approval is not itself a completed capital raise.
The next budget and the division of authority
Meta Compute is the initiative overseeing the construction and operation of Meta's AI infrastructure. Janardhan became a co-leader in January alongside Meta president Dina Powell McCormick and technical leader Daniel Gross, according to the transition account drawing on Bloomberg's reporting.
The accounts conflict about the technical role after the handover. Quartz, citing CNBC, says Brown will also serve as Meta Compute's chief technical officer. The Next Web says Janardhan will advise as its chief technology officer after the transition. Those descriptions leave the subsequent division of responsibilities unresolved.
The latter report also puts Meta's future commitments at about $825 billion, mostly for data centers and cloud computing, and says it has explored renting spare computing capacity to others, including talks over a $10 billion lease to prospective customer Anthropic. Those reported future commitments span a different scope from annual capital expenditures and cannot be added to the 2026 forecast. The leasing discussions do not establish a completed transaction or a realized source of revenue.
For Brown, the next consequential decisions concern how much capacity to secure and how to allocate it. Meta said its existing and scheduled capacity offered flexibility across model training, growth in its core business and new products. It claimed AI was already producing additional advertising revenue, while warning that its emerging AI investments would take longer to generate returns than investments in the core business.
A completed 2027 budget and a clarified Meta Compute leadership structure would show both the resources Brown will operate and the authority attached to his role. The appointment alone does not resolve how much of the expansion will serve Meta's own apps, support new AI products or become capacity sold to outside customers.
ByteDance’s video model is reaching creators through multiple platforms with 30-second 4K output. Posted prices and draft workflows show why the choice of distributor still matters.
Rox’s $1.3 billion valuation backs AI agents for large enterprise sales teams. Its growth and cost claims face a crowded market, while integration with existing systems leaves the case for replacing Salesforce unresolved.
Google’s Gemini Decision 1 scores predefined choices across text, images, audio, video and PDFs for $0.10 per million input tokens. Its shared-input billing targets recurring workflow costs, but cheaper decision models, rival caching and low-cost generative alternatives complicate the savings claim.
Treasury and other agencies reportedly want Singapore cloud provider Megaspeed on a U.S. trade blacklist. Commerce denies a reported cabinet confrontation over the delay, while Nvidia says it has blocked shipments since June 2025.
United is restricting unauthorized AI agents from searching, booking and managing travel through its platforms and intermediaries. The airline cites security and safety risks as it reserves control over future integrations.
A controlled study found that pharmaceutical ads shifted AI recommendations toward promoted drugs without reducing overall clinical accuracy. The results expose a gap in accuracy tests, but do not establish effects on prescriptions or patient outcomes.
Google is rolling out its Gemini app on Apple CarPlay, bringing voice conversations linked to Gmail and Google Calendar. It joins existing in-car AI options, while some personalized features remain restricted by age and region.
Google Labs’ Tabletop lets groups generate and use shared itineraries, polls and other interactive tools. The US adult-only experiment builds on an idea already offered by shareable AI app builders, leaving its practical advantage to be tested.
Anthropic’s IPO has reportedly attracted orders multiple times the available shares. Its planned $3 billion–$3.5 billion Japanese allocation would exceed Japan’s reported domestic IPO proceeds so far this year, but indications of interest are not completed sales.
Google warned that charging for news links would force it to stop distributing Indonesian news, while Meta and TikTok challenged other parts of the copyright bill. Existing Indonesian payment mechanisms and overseas deals complicate the platforms’ warnings about the cost to publishers.
Utrecht is investigating restrictions on camera glasses in municipal buildings, with findings expected in early 2027. The city must establish whether a ban is lawful, enforceable and compatible with assistive uses.
Research commissioned by cloud-storage supplier Wasabi estimates £3.8 billion in annual UK infrastructure overspend attributable to hard-to-forecast provider charges, including £1.2 billion for AI. Businesses report hiring and investment trade-offs, but recovering that money and turning it into jobs remain conditional.
The USDT issuer’s 800-million-parameter BotPsy beats a larger Qwen model on five of six company-reported agent benchmarks. Its local-AI ambitions still face missing handset measurements and explicit limits on sensitive uses.
Google DeepMind’s AlphaGenome 2 improves cell-specific DNA predictions and accepts new single-cell data without fine-tuning. Company benchmarks show gains at high confidence, but limited recall, complementary tools and greater computational demands leave experimental validation essential.
Coinbase is rolling out server-enforced trading limits and portfolio-loss triggers for AI agents connected to customer accounts. Product-category permissions remain forthcoming, and pausing new trades does not establish a ceiling on losses.
JetBrains IDEs 2026.3 bundle Air for running coding agents and reviewing their changes, while the standalone Air app is in maintenance mode. Existing provider subscriptions remain usable, but billing, security controls and execution options vary by integration.
Amazon’s shares and convertible notes could be worth $271.8 billion at the top of Anthropic’s proposed IPO range. Existing accounting gains, a 12% ownership cap and Anthropic’s computing commitments shape what that paper value means.
Nvidia chief Jensen Huang argued in May 2025 that US chip restrictions were strengthening Chinese rivals. Filings show a substantial commercial stake, but lost sales and market share alone cannot establish whether the controls failed their security purpose.
Reported demand has Anthropic’s bankers considering a price above its filed $70–$78 range. The combined financing could exceed $100 billion, but existing cash, uneven operating results and long-term computing commitments complicate the valuation test.
