Connect with us

Business

Early AI darling LangChain is now a unicorn with a fresh $125 million in funding

Published

2 months ago

on



LangChain, one of the earliest breakout startups of the generative AI era, announced a $125 million Series B funding round on Monday at a $1.25 billion valuation.

The startup, which created an eponymous open source framework for connecting AI apps to real-time data, hopes its tools can become the default building blocks that companies use to unleash a multitude of AI agents—while its investors believe the company has the potential to become as successful as other foundational digital infrastructure companies like Crowdstrike (for cybersecurity) and Datadog (for data monitoring). 

The round, which was rumored to have been completed over the summer, was led by IVP, with participation from existing investors Sequoia and Benchmark and new backers including CapitalG, Sapphire Ventures, ServiceNow Ventures, Workday Ventures, Cisco Ventures, Datadog, Databricks, and Frontline. LangChain says its tools are already used by AI teams at companies like Cisco, Replit, Clay, Cloudflare, Workday, and ServiceNow.

The company argues that building reliable AI agents—systems that can reason, act, and use tools on behalf of users—is still far too difficult. “Today, agents are easy to prototype but hard to ship,” LangChain wrote in a press release announcing the round. “Any input or change to an agent can create a host of unknown outcomes.”

The solution, the company says, is a new approach that blends product, engineering, and data science—what it calls agent engineering. The company is positioning itself as the connective tissue of the agent era—not just stitching together connectors, but providing the entire lifecycle of tools developers need to build, deploy, and monitor agents in production. A company like ServiceNow, for example, might use LangChain to connect an LLM to its internal knowledge base and use it to trigger workflows or track performance. 

LangChain began in late 2022 as an open-source project by Harrison Chase, then an engineer at Robust Intelligence, just weeks after OpenAI released ChatGPT. It pioneered the idea of “chains”—building blocks that connect large language models to external tools and data sources in a sequence, letting them take action instead of just generating text. A simple chain might let an AI take a user’s question, call a web search API, summarize the results, and return an answer—steps stitched together like links. It was an immediate hit:  “It was very crazy,” Chase recalled. “I didn’t know I was going to leave my previous job. I had no clue what I was going to do next.” 

It turned out that the project that became the startup LangChain, which Chase co-founded with Ankush Gola, became a darling of developers. That’s because it solved one of the most pressing problems in the early days of large language models: the models couldn’t access real-time information or perform actions like searching the web, calling APIs, or interacting with databases. LangChain’s framework let developers build those capabilities into their LLM apps—and adoption skyrocketed. The San Francisco startup raised a $10 million seed round led by Benchmark in April 2023, and announced a $25 million Series A in 2024 led by Sequoia, and valuing the company at $200 million.

Since then, however, the market has grown crowded with other companies offering similar tools, such as LlamaIndex and Haystack, while OpenAI, Anthropic, and Google now provide built-in capabilities that were once LangChain’s differentiators. 

To stay ahead, LangChain expanded its product lineup, including LangSmith, an observability, monitoring, evaluation and deployment platform built specifically for LLM applications and agents. Since launching last year, LangSmith has surged in popularity, as LangChain keeps some of its early products open source while creating proprietary platforms.

Langchain would not not provide details about its financials, thought a spokesperson said that a TechCrunch report in July that pegged its annual recurring revenue at between $12 million and $16 million was “low for where we are today.” While the company is not profitable, Langchain is “fairly efficient in spend” compared to high-growth, VC-backed startups, the spokesperson said.

IVP’s Tom Loverro, who led the investment, said the firm had “high conviction” in Chase and the company’s potential from the beginning. “Two years ago, the question was whether an open-source project like LangChain could become a major commercial company,” he said. “We saw Harrison and Ankush take the first important steps boldly into that journey,” including building multiple products that customers want. 

Loverro said he sees LangChain as potentially as successful as companies like Crowdstrike and Datadog, which became indispensable for taming the complexity of cybersecurity and cloud infrastructure, respectively. LangChain is betting it can become the layer that makes AI agents reliable and observable enough for enterprises to trust—turning today’s chaotic prototypes into business-critical systems.  “It feels increasingly sure that agents are super important to the future,” he said. “And if you believe that, then agent engineering is going to be incredibly important.” 

Chase admits the agent platform landscape is already crowded, but he argues LangChain’s breadth and neutrality will give it staying power. “There’s a ton of players,” he said. “I like to say we have 500 competitors and zero competitors at the same time.” Most enterprises, he predicts, will ultimately use multiple agent platforms, and many of them, like ServiceNow, will be powered under the hood by LangChain.

IVP’s Loverro emphasized that Langchain already has strong revenue, adoption, and big enterprises like Cisco and Workday building on LangChain. There will be competition, he says, “but it’s TBD if they matter.” And if the investors are right, LangChain could become the indispensable layer powering the agent era—just as CrowdStrike and Datadog did for the last generation of infrastructure.



Source link

Related Topics:
Continue Reading

Business

Rivian CEO says it’s a misconception EVs are politicized, with a 50-50 party split among R1 buyers

Published

7 minutes ago

on

December 10, 2025

By



If Rivian’s sales are any indication, owning an electric vehicle isn’t such a partisan issue, despite President Donald Trump’s rollbacks of mandates, incentives, and targets for EVs.

At the Fortune Brainstorm AI conference in San Francisco on Tuesday, Rivian CEO RJ Scaringe said it’s a misconception that electrification is politicized, explaining that most customers buy a product based on how it fits their needs, not their ideology. The questions car buyers ask, he said, are the same whether they’re purchasing one with an internal-combustion engine or a battery: “Is it exciting? Are you attracted to the product? Does it draw you in? Does the brand positioning resonate with you? Do the features answer needs that you have?”

Buyers of Rivian’s R1 electric SUV are split roughly 50-50 between Republicans and Democrats, Scaringe told Fortune’s Andrew Nusca. “I think that’s extraordinarily powerful news for us to recognize—that this isn’t just left-leaning buyers,” he added. “These are people that are saying, ‘I like the idea of this product, I’m excited about it.’ And this is thousands and thousands of customers. This is statistically relevant information.”

Buying an EV was once an indication of left-leaning politics, but the politics got scrambled after Tesla CEO Elon Musk became the top Republican donor and a close adviser to Trump. That drew some new customers to Tesla, and turned off a lot of progressive EV buyers, with many existing owners putting bumper stickers on their Teslas explaining that they bought their cars before Musk’s hard-right turn. Trump and Musk later had a stunning public feud, in part over the administration’s elimination of EV and solar tax credits.

But Scaringe said he started Rivian with a long-term view, independent of any policy framework or political trends. He also insisted that if Americans have more EV choices, sales would follow. Right now, Tesla dominates a key corner of the market, namely EVs in the $50,000 price range. Rivian’s forthcoming R2 mid-size SUV will represent a new choice in that market, with a starting price of $45,000 versus the R1’s $70,000.

Ten years from now, Scaringe said he hopes—and believes—that EV adoption in the U.S. will be meaningfully higher than it is today across the board, explaining that the main constraint isn’t on the demand side. Instead, it’s on the supply side, which suffers from “a shocking lack of choice,” especially compared to Europe and China, he added. EV options in the U.S. are limited by the fact that Chinese brands are shut out of the market.

More choices for U.S. EV buyers would presumably create more competition for Rivian—and indeed, the flood of low-priced Chinese EVs in other auto markets has created a backlash, with countries such as Canada imposing steep tariffs on them. But Scaringe appears to view more competition as positive for the market overall.

“I do think that the existence of choice will help drive more penetration, and it actually creates a unique opportunity in the United States,” he said.



Source link

Continue Reading

Business

Powell warns of a ‘very unusual’ economy as inflation remains high amid a weakening job market

Published

38 minutes ago

on

December 10, 2025

By



Federal Reserve Chair Jerome Powell on Wednesday described the U.S. economy as “very unusual,” saying policymakers are navigating a rare combination of tariff-driven goods inflation and a labor market that may already be weaker than official data suggests.

The Fed cut interest rates for the third consecutive meeting, a quarter-point reduction Powell framed not as a confident pivot toward easier policy, but as a defensive move meant to keep the labor market from slipping further. He repeatedly emphasized risks to employment have risen “in recent months,” and noted that behind the headline numbers, job creation may already be negative.

Powell made the striking admission the Fed believes the official payroll figures—which have slowed sharply since the summer—are overstating job growth by roughly 60,000 per month. 

“Forty thousand jobs could be negative 20,” he said, adding this dynamic is not well understood by the public because unemployment claims remain historically low—something both economists Mark Zandi and Claudia Sahm recently toldFortune could be giving people a false sense of security about the job market.

“I think a world where job creation is negative… we need to watch that very carefully,” Powell said. 

It is this weakening backdrop Powell said makes the current moment “very unusual”: Inflation remains elevated, but most of the remaining overshoot comes from goods categories directly affected by tariffs, as opposed to domestic economic overheating, which he said the Fed has worked hard to cool since its 2022 highs; inflation excluding tariff-affected goods is “in the low [two percent],” he said. Services inflation is cooling, wage pressures are easing, and neither the labor market nor business surveys suggest a “Phillips-curve” kind of inflation threat, Powell said, referring to the inverse relationship between inflation and unemployment. 

Instead, Powell said, the bulk of the problem is a “one-time price increase” pushing up goods categories as import levies work their way through supply chains. Goods inflation, he noted, should peak around the first quarter of 2026, assuming no additional tariff rounds.

Those crosscurrents have fractured the Fed. Three officials formally dissented from the rate cut on Wednesday, and several others offered what Powell described as “soft dissents,” when an official’s personal projection falls out of what they ultimately voted for. There were six such “soft dissents” this time, during one of the deepest divides inside the FOMC in years, driven by disagreement over how to weigh the risks of lingering inflation against the possibility that job growth is weaker—and much more fragile—than reported.

Powell stressed that policymakers cannot simply choose one mandate to prioritize. 

“There is no risk-free path,” he said, a refrain he’s repeated for months. “When both sides of the mandate are threatened, you should be kind of neutral.” 

He characterized the current stance as being at the “high end” of neutral, allowing the Fed to “wait and see” how the data evolve.



Source link

Continue Reading

Business

Top economist Diane Swonk: Jerome Powell risks losing the Fed’s credibility on a gamble about AI and immigration

Published

1 hour ago

on

December 10, 2025

By



Federal Reserve Chair Jerome Powell warned Wednesday afternoon that the U.S. labor market may be significantly weaker than the official data suggest. But according to KPMG chief economist Diane Swonk, the Fed may be drawing the wrong conclusion—and in doing so, risks undermining its hard-won credibility on fighting inflation.

In a new analysis shared with Fortune, Swonk argues that Powell is treating the slowdown in hiring as a sign of weakening demand that must be offset with lower interest rates. But if that weakness is being driven instead by structural forces—specifically, AI adoption and sharp declines in immigration—then cutting rates won’t fix the underlying problem and could worsen inflation.

“Powell risks the Fed’s inflation-fighting credibility if the weakness in employment is due more to AI and curbs in immigration than weak demand,” Swonk wrote.

That warning comes after one of the most contentious Federal Open Market Committee meetings in years. The Fed cut rates by a quarter point for the third meeting in a row, taking the federal funds rate down to 3.5%–3.75%, but the vote fractured the committee. Swonk notes it was the first time since 2019 that there were three dissents, and they came “in opposite directions.”

Governor Stephen Miran — currently on leave from the White House Council of Economic Advisers — voted for a half-point cut, while Kansas City Fed President Jeff Schmid and Chicago Fed President Austan Goolsbee voted to hold rates steady.

Swonk highlights that the Fed’s statement resurrected language meant to indicate a pause: “In considering the extent and timing of additional adjustments… the Committee will carefully assess incoming data, the evolving outlook and the balance of risks.” Powell reinforced that stance, saying “We are well positioned to see how the economy evolves” and emphasizing that policymakers would need to “be a bit skeptical” of data distorted by the government shutdown.

But the bigger issue, Swonk argues, is that Powell kept pointing to imminent downward revisions to employment, revisions she warns may not mean what the Fed thinks they do.

If job growth is negative because automation is replacing workers or because the labor force is shrinking due to immigration policy, then monetary policy can’t solve the problem. That’s because rate cuts can stimulate demand, but they cannot create workers or reverse automation decisions already made by firms. 

“The challenge is if that weakness is due to AI and curbs on immigration, then rate cuts will not do much to shore up the labor market. More could show up in inflation,” she wrote.

Powell, during the conference, acknowledged that AI may be “part of the story” behind the cooling labor market, citing major employers like Amazon that have linked hiring freezes and job cuts to automation. But he stressed that it’s “not a big part of the story yet,” and said it’s too early to know whether this wave of technological change will ultimately destroy more jobs than it creates.

He also noted that labor supply has “come down quite sharply” due to a drop in immigration and participation.

A misread could become especially dangerous given the fiscal backdrop. Swonk notes that “expansions to tax cuts last year will show up as a record high tax refunds in early 2026,” warning that the windfall could “further entrench inflation much like we saw in the wake of the pandemic.” 

At the same time, federal debt is projected to surpass GDP for the first time since World War II, marking a level of issuance that is “a lot of debt for bond markets to absorb.”

Swonk also flags mounting risks to credibility inside the Fed itself.

Six participants wanted to hold rates steady, and the market openly dismissed Powell’s attempt at a hawkish spin: investors “priced in more cuts after the meeting,” she notes. Powell now appears to be one of the more dovish voices on the committee, raising questions about the direction of policy if the administration installs a new chair aligned with Miran’s more aggressive easing stance.

Swonk expects the Fed to pause early next year, but warns that if inflation fails to cool as expected, “the bond market could grow more skittish about rate cuts.”



Source link

Continue Reading

Trending

Copyright © Miami Select.