Start-up Anysphere Raises $105 Million in Funding

Anysphere, whose Cursor software uses artificial intelligence to help programmers code on the fly, plans to announce on Tuesday that it has raised $105 million at a big jump in valuation.

Funding and Valuation

The new round, which values the nearly three-year-old start-up at $2.5 billion, was led by Thrive Capital, which spearheaded huge recent investments in OpenAI and the data company Databricks, and Andreessen Horowitz, the giant venture capital firm that is another existing backer.

The funding is the latest sign of fierce competition among investors to gain exposure to A.I. start-ups. A.I. deals accounted for more than 35 percent of the $366 billion worth of venture capital deals last year, according to the research firm PitchBook.

About Anysphere and Its Product

A.I. start-ups are hot, with Anysphere being no exception. The company’s Cursor software operates off proprietary models and large language models from OpenAI and Anthropic, helping to autocomplete code for programmers.

On Monday, the company unveiled a new version of a feature that can predict a user’s next line of code — with its CEO saying that in some cases it can fill in the next 10 minutes’ worth of programming.

New Features and Adoption

Its something that many tech moguls have high hopes for: Mark Zuckerberg, the chief executive of Meta, said recently on “The Joe Rogan Experience” that his company will begin automating some code-writing.

Anysphere has also quietly rolled out so-called agentic features for Cursor that can independently complete some tasks for users. “Source code is starting to blend into the background,” its CEO said.

Competition and Growth

Other companies, including Codeium, Poolside and Sourcegraph, offer competing products, and the Copilot feature of Microsoft’s GitHub also focuses on automating coding. In November, Anysphere bought another competitor, Supermaven.

Developers from all of Thrive’s portfolio companies use Cursor, up from a handful a year ago. This was part of what made Thrive’s partner receptive to Anysphere’s argument in the fall of 2024 that the company should raise money again, given its growth.

Concerns and Conclusion

For some skeptics, the investor A.I. fervor — especially how quickly it has pushed up funding rounds — has increased the possibility of companies being overvalued, as in previous technology booms.

However, Anysphere’s CEO argued that his company had sought to keep a lid on costs, including spending on computing resources. He also argued that his company’s valuation was within historical norms. (It’s about 25 times Anysphere’s $100 million in annual recurring revenue, in line with other A.I. start-ups.)

Conclusion: Anysphere’s new funding and valuation are a testament to its rapid growth and adoption. While there are concerns about valuations in the A.I. start-up space, Anysphere’s CEO and investors believe that its product and business model are solid.

Frequently Asked Questions

Q: What does Anysphere do?

A: Anysphere is a start-up that develops software using artificial intelligence to help programmers code on the fly.

Q: What is Anysphere’s new valuation?

A: Anysphere’s new valuation is $2.5 billion, after raising $105 million in a new funding round.

Q: What is Anysphere’s product, Cursor, used for?

A: Cursor is used to help programmers code by predicting their next line of code and providing suggestions for completing tasks.

University students from all backgrounds are encouraged to access highly competitive roles post-graduation by demonstrating their experience in the field. But how can this be possible if they are in full-time education? The solution: internships. Internships are defined by the UK government’s Department for Business and Trade as ‘positions requiring a higher level of qualification than other forms of work experience, and are associated with gaining experience for a professional career’. However, as internship activity is unregulated, there is no single definition covering what constitutes an internship. Additionally, an internship can be of any duration up to one year and relies on an organisation to design an opportunity that is of benefit to the student. Critically, internships are external to any qualification such as a placement year, typically sourced independently by students, and are completed around their studies without financial and wellbeing support from the institution.

Young people from working-class backgrounds already face systemic barriers to accessing internships, and it is worth exploring what impact Labour’s plan to ban unpaid internships in its ‘Delivering a New Deal for Working People’ manifesto might have on them. Although welcomed by many, this amendment has the potential to both positively and negatively impact students from lower socioeconomic backgrounds and contribute to further structural inequalities in the employment relationship.

Many governmental bodies use the term ‘level the playing field’ and insist that as a nation we in the UK live in a meritocratic society wherein the ability to secure an affluent job role is based on effort and outcome rather than on the social networks that are implicitly ingrained in all that we do.

However, this meritocratic narrative is inaccurate for two important reasons. First, many studies suggest that people from working-class backgrounds are at a pre-existing disadvantage due to limited networks in their chosen fields, minimal or no financial or social support from family, and reduced geographic mobility. Additionally, they face multiple barriers to higher education because of a system that is inherently biased from early childhood, offering starkly contrasting education standards in private, grammar and state schools in the UK. The Institute for Fiscal Studies 2023 report suggested that social mobility was at its lowest point in over 50 years.

Second, biases — be they explicit or implicit — are prevalent in the workplace in the form of ‘cultural fit’. These biases can arise from the insider jokes and stories told among colleagues, the clothing and brands that you wear, where and with whom you went to school, your geographic location, your hobbies, the car you drive and all of the implicit markers that make you an individual alongside the protected characteristics under UK law.

For protection, human beings instinctively gravitate towards those who are similar in personality, attitude or background, thus creating a level of trust among peers. In most social situations this helps us create friendships, overcome adversity and thrive. However, these subconscious biases become problematic when managers hire in their own image, thus creating the monotypic non-diverse CEO boards that many organisations are trying to change.

An intern’s ‘success’ in recruitment, experience and future career development can be based on this cultural fit within their target organisation. Some biases persist even if actors within organisations use blind-screening processes or artificial intelligence to reduce familiarity biases. These biases can be from the level of education on an individual’s CV or the standing of some universities, which many ‘elite’ organisations use as a benchmark for any level of recruitment access.

Although most Russell Group universities actively recruit students from ‘widening participation backgrounds’, working-class individuals will typically have faced systemic barriers to accessing these universities through their primary and secondary education and therefore are already put at a disadvantage by their attendance at another university, regardless of merit. These barriers arise from the level of educational support through public education and funding, and a lack of understanding and insider support from family and friends who may not know how to ‘play the game’ or present in a certain way that portrays ‘fit’ to prospective employers.

But aren’t internships created equally in design? The short answer is no. Internships may be paid, unpaid, designed around a structured programme, or used as an employee substitute — although this is technically unlawful. There is also an unspoken but accepted hierarchy of internships: Those that are paid are perceived to offer more meaningful opportunities as they are thought to contribute to the business; unpaid internships, on the other hand, are recognised as admin-based or involving tea- and coffee-making, although the reality may be very different. As with any employment opportunity, paid internships are scarce and therefore highly competitive, leading to those with the cultural fit or who can evidence the symbolic capital on their CV — through personal and family work experiences in the industry – securing these limited roles.

Students from lower socioeconomic backgrounds may then seek to explore unpaid internships due to the wider selection of opportunities, but find them unsuitable if they need to support their families or themselves. As a result, working-class students regularly return to lower-skilled, paid part-time service roles, instead of the employability-building opportunities that are encouraged to develop that graduate identity.

If unpaid internships are banned under the Labour government, this may reduce some exploitative opportunities; however, the policy could also reduce the number of roles available, and result in paid internships becoming even more competitive, with those with symbolic merit and cultural ‘fit’ securing even more frequently those career-enhancing opportunities.

In addition to the hierarchy of paid and unpaid internships, we also now see a hierarchy of virtual versus in-person internships. Virtual internships are thought to be of less benefit to the individual because they are not grounded in the ‘office’ where students can experience the ‘true’ culture of the work. Some students from lower socioeconomic backgrounds may be encouraged to complete virtual internships, because they are less financially able to relocate to city-based firms. In contrast, those with personal or family wealth and networks are able to complete these more affluent in-person opportunities – leading to a greater likelihood of securing an appropriate full-time role post-graduation.

Labour, in its January 2024 ‘Plan to Make Work Pay’ manifesto, stated that it would ‘ban unpaid internships except when they are part of an education or training course’. As of the last employment bill amendment that was put forward to parliament on 10 October 2024, this change in legislation was yet to be included. The Sutton Trust, which campaigns for socioeconomic equality, has continually called for unpaid internships of over four weeks in duration to be banned. Neither amendment is yet to be enacted by this or previous governments.

There are many systemic layers relating to the unprotected characteristic of social class which leads to biases in the recruitment and retention of individuals from class-marginalised backgrounds. A ban on unpaid internships could have both a positive impact as well as a negative one. Regardless of whether this ban may come into force, businesses should take Labour’s inclusion of this amendment as a ‘red flag’ to consider socioeconomic class barriers in their diversity metrics. Human resource teams must consider how they can open up both internships and subsequent graduate roles to a wider group of candidates, in a way that encompasses the holistic lived experience of a candidate, rather than using a tick-box approach to assess the cultural fit of their CV.

Naomi Wells is a PhD researcher and associate lecturer within the Work Informalisation and Place Research Centre at Nottingham Trent University. Her research focuses on the impact of precarity on social mobility.

Image credit: Štefan Štefančík via Unsplash

Far Above Your Head: The Constellations Working to Keep Us Safe

Far above your head, constellations of satellites are working constantly to provide the positioning, navigation, and timing systems that quietly run modern life.

The Threat: GPS Interference

GPS signals can be jammed—deliberately drowned out with other powerful radio signals—and spoofed, where erroneous signals are released to fool positioning systems. GPS interference has been documented in Ukraine, the Middle East, and the South China Sea.

The Solution: Artificial Intelligence and MagNav

But startup SandboxAQ believes that artificial intelligence, when combined with navigation systems that read Earth’s magnetic field, known as MagNav, could mitigate these threats to GNSS. “Our technology does not replace [GNSS], but can enrich existing navigation systems to improve safety and serve as an alternative primary navigation source in case of GPS outages,” says Luca Ferrara, general manager of SandboxAQ’s navigation department.

How it Works

SandboxAQ’s navigation technology, called AQNav, uses quantum magnetometers—devices that can detect changes in magnetic fields very precisely by measuring subatomic particles—to produce a reading of the Earth’s magnetic field. “We are looking for the unique fingerprint of magnetized rock formations in the Earth’s crust,” says Ferrara.

Artificial intelligence is then used to accurately pinpoint the position of an aircraft, through comparison with known maps of the magnetic field. The AI also eliminates any external interference generated by the aircraft, such as from sudden movements or signals from its electrical systems; individual aircraft have unique characteristics when it comes to introducing magnetic interference.

Testing and Deployment

So far, the US Air Force, Boeing, and Airbus have all made test flights using the system. “Since May 2023, we have deployed and tested many iterations of our hardware and software,” Ferrara explains. “AQNav has flown hundreds of kilometers in different types of aircraft, from single-engine aircrafts to large military transports. It has been tested in real flight scenarios, including two major military exercises by the US Air Force.”

Conclusion

The reliable and secure provision of navigation and positioning systems is crucial for modern life. However, the threat of GPS interference is real and growing. SandboxAQ’s AQNav technology, combining artificial intelligence and MagNav, offers a solution to this problem, providing a reliable and alternative primary navigation source in case of GPS outages. With successful testing and deployment by major players in the aerospace industry, AQNav is poised to revolutionize the way we navigate.

FAQs

Q: What is the purpose of SandboxAQ’s AQNav technology?
A: AQNav is a navigation system that uses artificial intelligence and MagNav to provide a reliable and alternative primary navigation source in case of GPS outages.

Q: How does AQNav work?
A: AQNav uses quantum magnetometers to detect changes in magnetic fields, and artificial intelligence to pinpoint the position of an aircraft through comparison with known maps of the magnetic field.

Q: Who has tested and deployed AQNav?
A: The US Air Force, Boeing, and Airbus have all made test flights using the system.

Q: How effective is AQNav?
A: AQNav has flown hundreds of kilometers in different types of aircraft, from single-engine aircrafts to large military transports, and has been tested in real flight scenarios, including two major military exercises by the US Air Force.