|
Welcome back! Software companies have been at pains to flaunt the success of their AI products, as they seek to dispel investor fears that AI apps are threatening their existence. Two reports this week show that developing popular AI products is only half the challenge. Canva, the design software company popular among students and business owners, warned investors this week that its annual revenue growth would slow to 20%. The reason: it was tapping the brakes on a planned AI rollout expected to drive paying subscriptions—after demand had proved higher and more expensive to serve than anticipated. Then on Thursday, Figma said it expects a slowdown in revenue growth for the September quarter, to 36% from 48% in June. In its case, the publicly traded design software company suggested it’s still figuring out how to generate sales from new AI tools that are still in beta, or testing phases, or early in their release. Both companies are cognizant that any success they have with AI comes at a potentially high cost. Canva COO Cliff Obrecht told my colleagues that before AI, the cost of serving its many free users was “very low.” With AI, “those costs became much higher. The unit economics changed, which made it far more important for us to bring those AI costs down,” he said. Figma, which largely sells subscriptions to corporate clients, says the costs are likely to hit its gross margins. “We do not charge our customers for their usage of products that are currently in beta, and we bear the cost of inference without offsetting consumption revenue,” said CFO Praveer Melwani in a call with investors. One solution for both is to lower costs with models they train themselves. Canva is heavily pushing its own in-house AI model it says is much cheaper and faster at creating images and videos than top models from AI labs. But these weren’t ready in time for the rollout of its AI features. Figma executives said the company is also investing more in its in-house AI models. It’s started to ramp up use of these alongside frontier models to run its new Figma AI agent. But training a model also takes time and money—and investors are an impatient lot. Figma shares fell about 15% Thursday after the release. Only a Sliver of Forward Deployed Engineers Are Delivering AI Returns, Study Finds Only a fraction of the roughly 17,000 tech consultants working in the U.S. with the title of forward deployed engineers are helping businesses or government agencies see returns from AI agents, according to a new study from executive recruiting firm Christian & Timbers. The job of forward deployed engineer—a hybrid role that’s part-software engineer, part business consultant—has been all the rage in 2026. Pioneered by Palantir, this year companies including Meta, Google, Amazon, Salesforce, and new ventures started by the likes of Anthropic and OpenAI have ramped up efforts to hire people for FDE roles aimed at helping businesses use their AI products. But Christian & Timbers found that, so far, only about 12% of FDEs have helped several enterprises deliver 10 million or more in cost savings or revenue, by building custom AI agents. Christian & Timbers interviewed more than 250 C-suite hiring executives and 300 forward deployed engineers across 180 companies, the majority of which were large enterprises, and compiled data from LinkedIn and Indeed. It’s another data point suggesting that enterprises are still in the early stages of figuring out how to get returns from their spending on in artificial intelligence. Whether or not they succeed could help determine whether enterprises spend big sums on AI firms that themselves need to see a big return on their own investments in the tech. Christian & Timbers CEO Jeff Christian said that which companies hire the best FDEs will help determine how successful their AI products are: “All that matters is how you get them to work and automate stuff….it’s the deployment now that matters.” The overwhelming majority of the so-called group of “elite” FDEs, or about 80%, are employed by Palantir. The others work for Anthropic, OpenAI’s DeployCo, Google, Amazon, and Microsoft, Christian told me. Typically those high-level FDEs specialize in working in one particular sector, such as semiconductors or financial services. Of the other FDEs, most are early career engineers making between $175,000 and $300,000 in compensation. Palantir CEO Alexander Karp has been salty about the industry copying its forward deployed engineering model. In late July, Karp said at a customer event: “For those of you who are too young to remember, the concept of FDE was derided by everyone in Silicon Valley…derided by every company that’s now copying it.” Demand for FDEs has surged, Christian & Timbers found. While only 10% of the companies it studied had planned to hire FDEs at the start of the year, approximately 70% had hired or planned to hire FDEs by the end of the June quarter.
| 
