Beyond Elite Education: How Families Navigate Access to the World's Most Selective Careers

Executive summary

Families invest heavily in education on the reasonable assumption that strong credentials open doors. The evidence supports part of that assumption. Elite schools and universities are heavily over-represented among those who reach the top of British and American professional life, and the strongest causal study now available finds that attending an Ivy-Plus college nearly triples a student's chance of working at a prestigious firm. But a credential is an entry ticket to a separate contest, not the outcome of it.

This article sets out what the evidence says about that second contest: the selection systems run by investment banks, private capital firms, corporate law firms and strategy consultancies. Drawing on the Who Gets In? register of 1,387 dated observations from 494 official employer and university sources, it finds three features that families routinely underestimate. First, selection is multi-stage and stage-specific: academic screens, timed tests, recorded interviews, case work and assessment centres each test something different. Second, the clock runs early: for many sought-after routes, the decisive application falls in the first or second year of university, often 18 months or more before a graduate start date. Third, the rules are uneven and partly invisible: of 67 selective employers checked in October 2026, no official statement on AI in selection was located for 30, including several of the most sought-after banks and every private equity, hedge fund and trading firm checked. Not located is not the same as no rule.

The article proposes an analytical model separating educational attainment, career information, recruitment eligibility, selection readiness and professional opportunity. Evidence is strong at the two ends of that chain and weak in the middle, especially on what families know and when. It does not claim that elite education fails, that families need private advice, or that AI is transforming selection. It identifies where the evidence runs out, and sets a research agenda that professional advisers can help answer.

The scale of educational investment

The investment families make before a child applies for a first job is substantial, and in the United Kingdom it has recently risen sharply. The Independent Schools Council reported that the average day school fee in January 2025 was GBP 7,382 a term including VAT, 22.6 per cent higher than a year earlier, after VAT was applied to fees from 1 January 2025 [ISC 2025]. Boarding costs considerably more. Outside the fee-paying sector, private tutoring has become widespread: the Sutton Trust's 2026 polling found that 29 per cent of secondary school pupils in England and Wales had received private tuition at some point, up from 18 per cent in 2005 [Sutton Trust 2026].

For internationally mobile families, the investment often extends further: overseas schooling, admissions preparation for UK or US universities, and in many cases a postgraduate degree. We were not able to locate a reliable estimate of total family spending on the education-to-career pathway across our five priority markets, and we do not offer one.

What this investment is meant to buy is usually unstated. For many families the implicit model is linear: a strong school leads to a strong university, which leads to a strong career. The rest of this article tests the third link.

Evidence status. Fee and tutoring figures are published survey data (fact). The linear model is our characterisation of a common assumption, not a measured belief (hypothesis).

Why educational attainment and professional selection are different

The association between elite education and elite careers is not in doubt. In the United Kingdom, Elitist Britain 2019, a study of more than 5,000 people in leading positions, found that 39 per cent had attended independent schools, compared with 7 per cent of the population [Sutton Trust and SMC 2019]. In the United States, Chetty, Deming and Friedman, using admissions records linked to tax data and a design that exploits idiosyncratic waitlist decisions, estimate that attending an Ivy-Plus college rather than an average highly selective public flagship raises a student's chance of reaching the top 1 per cent of earners by about 50 per cent, nearly doubles the chance of attending an elite graduate school and almost triples the chance of working at a prestigious firm [Chetty, Deming and Friedman 2026].

Three qualifications matter for families.

First, correlation is not the same as causation, and the causal effect varies. Dale and Krueger, comparing students admitted to similar sets of colleges, found that the earnings premium for attending the more selective college generally fell to close to zero once unobserved student characteristics were accounted for, with notable exceptions for Black and Hispanic students and those from less-educated families [Dale and Krueger 2014]. Chetty and colleagues reconcile their findings with this work: the effect of an elite college is concentrated in the upper tail of outcomes, including prestigious employment, rather than in average earnings.

Second, the credential does not complete the job. The UK Social Mobility and Child Poverty Commission's study of 13 elite law, accountancy and financial services firms found that entry remained dominated by people from privileged backgrounds, which it attributed primarily to recruitment from a narrow group of universities, and that firms defined talent partly through confidence and "polish", qualities participants acknowledged could map onto middle-class socialisation [Ashley et al. 2015]. In the United States, Lauren Rivera's study of hiring in elite banks, law firms and consultancies, based on 120 interviews with evaluators, found that hiring was also a process of cultural matching between candidates, evaluators and firms [Rivera 2012].

Third, advantage persists after entry. Using the Labour Force Survey, Laurison and Friedman found that people from working-class origins in higher professional and managerial occupations earned on average 16 per cent less than colleagues from privileged backgrounds [Laurison and Friedman 2016]. Research for the City of London Corporation found that employees from lower socio-economic backgrounds in eight financial services firms took about 25 per cent longer to progress, a gap not explained by performance [Bridge Group 2020, as reported in Sutton Trust 2021].

Read together, these studies support a distinction that families do not always draw. An elite education raises the probability of being considered. It does not determine what happens inside a selection process that tests different things, at different times, in different ways.

Evidence status. The association and the Ivy-Plus effect on prestigious employment are established by peer-reviewed research (fact). That families conflate admission with selection is a hypothesis this article cannot yet test with survey evidence.

How leading employers actually assess candidates

The Who Gets In? register records how selective employers and universities describe their own selection processes. Each row records one observation from one official source for one institution, programme, market and date. As read on 9 November 2026, the register held 1,387 signals from 494 official source records, last updated on 2 November 2026 [WGI register]. It is a purposive collection weighted towards London, banking and law. It cannot estimate how common any practice is, and nothing below should be read as describing all employers.

Four features stand out.

Selection is a sequence, not a judgement. A typical published process combines an application, one or more timed or untimed online assessments, a recorded or live interview, and a final assessment centre or "Super Day". A&O Shearman's London training contract, for example, publishes an online application and an online experience of situational judgement, verbal and numerical reasoning, followed by two one-to-one interviews [WGI SIG-185, SIG-186]. Baird's London internship invites shortlisted candidates to a first-round interview and then a Super Day with motivational, technical and conversational elements [WGI SIG-364, SIG-365]. Each stage is a separate filter, and each filters on something different.

Academic screens vary more than families might expect. Some employers publish firm thresholds: A&O Shearman's London training contract lists AAB at A-level, a 2:1 and seven GCSEs at grade 6 or above [WGI SIG-284, SIG-294, SIG-295]. Others publish none: Ashurst states no minimum A-level requirement [WGI SIG-327], and Baird states that no prior industry experience is needed for its London summer internship [WGI SIG-362]. A Bain global vacancy refers to a degree from "a renowned academic institution" [WGI SIG-355], while A&O Shearman's London page states that graduates of all universities are welcome [WGI SIG-299]. Prospects Luminate reports that the proportion of graduate employers requiring specific A-level grades fell from 26 per cent in 2015/16 to 13 per cent in 2024/25 [Prospects Luminate 2025]. A strong academic record is necessary for most of these routes; it is rarely sufficient, and the screens that matter differ by employer.

Application rules constrain strategy. Several employers limit how often a candidate may try. Barclays permits one full-time graduate or internship application globally each year [WGI SIG-377]. A&O Shearman permits one opportunity per cycle in London and requires candidates who fail its online experience to wait for the next cycle [WGI SIG-184, SIG-289]. BCG allows only one office to progress an application at a time and normally expects about a year before a same-office reapplication [WGI SIG-1171, SIG-1173]. BlackRock gives five days to complete its pre-interview assessment, after which an application is withdrawn [WGI SIG-128, SIG-1183]. A candidate who is academically exceptional but unaware of these rules can lose a cycle to a procedural error.

The same employer may run different processes by market. BlackRock's 2027 EMEA Client and Product Summer Internship lists a deadline of 30 September 2026 for London and 31 January 2027 for Abu Dhabi and Riyadh [WGI SIG-126, SIG-1246]. A&O Shearman's Singapore summer vacation scheme is the only formal trainee route there [WGI SIG-1048], and its Hong Kong scheme sets language preferences that its London scheme does not [WGI SIG-1095]. Families moving between markets cannot assume that one office's process describes another's.

Why might an academically exceptional candidate be poorly prepared for a given process? On the evidence above, the answer is structural rather than personal. University assessment rewards sustained work on known material over weeks. Employer selection, as published, rewards timed reasoning, spoken explanation under observation, commercial judgement in case form and procedural compliance with rules that change by employer and market. These are different capabilities, and nothing in the published evidence suggests that strength in the first guarantees strength in the second.

Evidence status. Each example is an employer's own published statement, observed on the date recorded (fact). The claim that university assessment and employer selection test different capabilities is our interpretation of those published formats (inference). Whether published processes match practice is not established.

Why recruitment timing matters

The clearest gap between what families may expect and what employers publish concerns timing.

In UK corporate law, A&O Shearman's Summer 2027 London vacation scheme is open to penultimate-year students and closed on 9 November 2026 [WGI SIG-292, SIG-298]. All participants are considered for a training contract [WGI SIG-290]. A first-year programme can fast-track participants to the vacation scheme assessment centre [WGI SIG-019, SIG-286]. For a student on a three-year degree, the most important application may therefore be made in the first term of the second year, for work beginning roughly two years later.

In banking, Bank of America's London 2027 Summer Analyst vacancy states that interns who perform well may be offered full-time roles [WGI SIG-370], and that assessments often begin before the published deadline [WGI SIG-376]. Its spring insight programme is aimed at first-year undergraduates [WGI SIG-372]. BlackRock's 2028 APAC Full-Time Analyst listing, already observed in October 2026, describes a graduation window running from September 2026 to August 2028 [WGI SIG-1180].

In the United States, the cycle runs earlier still. Yale's Office of Career Strategy publishes guidance on 2028 summer internships addressed to sophomores [Yale OCS 2026]. In 2018, Goldman Sachs and JPMorgan announced that they would stop interviewing sophomores for certain internships [Daily Pennsylvanian 2018]; current university guidance indicates that sophomore-year recruiting has since returned, but we have not located an official employer statement establishing the current calendar across banks.

Internships matter because they function as extended selection. In the United States, NACE reported an average intern-to-full-time conversion rate of 63.1 per cent for 2024/25 interns, the highest in five years [NACE 2026]. In the United Kingdom, 77 per cent of employers in ISE's development survey agreed that former interns were more work-ready [ISE 2025b]. Meanwhile, competition for the opportunities themselves has risen: ISE reported that applications for university-level internships rose 27 per cent in 2025/26 [THE 2026].

Private capital is different. Apax described in 2020 an analyst programme hiring directly from universities, while stating that most of its investment team joined at associate level [WGI SIG-316, SIG-317]. Blackstone's student programmes are aimed at juniors and final-year students [WGI SIG-390, SIG-391]. For many private capital routes, the relevant decision is therefore not only which internship to win but which first employer, often a bank or consultancy, will make a later move possible. We have not found systematic evidence on how private capital firms weigh prior employers, and we treat this as an open question.

Evidence status. Deadlines, eligibility windows and conversion routes are published employer statements (fact). Conversion rates are employer survey data with disclosed samples (fact, US and UK respectively). That families underestimate timing is a hypothesis: we found no survey measuring parental awareness of recruitment calendars.

Access, inequality and information

Two bodies of evidence meet here. The first shows that selection advantages persist beyond credentials: in recruitment from a narrow set of universities, in the reward for "polish", in cultural matching and in progression after entry (see why educational attainment and professional selection are different). The second shows that preparation for selection is itself unequally distributed. Private tutoring, the most visible form of paid preparation, is more common among better-off families [Sutton Trust 2026].

Information is a third and less studied channel. Every rule described in the section on how leading employers actually assess candidates is public, but it is scattered across hundreds of pages, written for different markets, and revised each cycle. The Who Gets In? AI Rules Tracker illustrates the point. Of 67 selective employers checked on 23 October 2026, 37 published an official statement on AI in selection. Of the 31 that addressed candidates' own use, 25 allowed some AI help with written applications and none forbade it; 18 forbade AI in interviews and none allowed it [WGI AI Rules Tracker v2.1]. But no statement was located for 30 of the 67, including Goldman Sachs, JPMorganChase, Morgan Stanley, Linklaters and Freshfields, and every private equity, hedge fund and trading firm checked. "Not located" means not found on public official pages on the observation date; rules may be given privately to candidates.

This creates an information gradient. Where rules are public, any candidate can read them. Where they are communicated privately, through portals, recruiter emails or word of mouth, access depends on networks. We did not measure whether that gradient affects outcomes, but it is consistent with the qualitative findings of Ashley and colleagues on support available to students at targeted universities [Ashley et al. 2015].

Evidence status. Inequality in access and progression is well established (fact). The information gradient is our inference from the pattern of public and missing statements; its effect on outcomes is untested.

The overlooked question of next-generation family advisory

Wealthy families already receive structured advice about the next generation, but the published agenda is mainly about wealth. UBS's Global Family Office Report 2025, based on 317 single family offices, found that 53 per cent had wealth succession plans; among those with plans, 43 per cent saw preparing the next generation to take on wealth responsibly as a challenge, and only 26 per cent consulted the next generation from the outset [UBS 2025]. Several private banks run next-generation programmes. Public descriptions of those we reviewed emphasise financial education, leadership, family business and philanthropy [Citi Private Bank; Spear's 2026; Private Banker International].

We found no published research on how often families raise a child's professional career with their private banker, family office or lawyer, or what those advisers do when they do. That gap matters for three reasons. First, the timing problem in the section on why recruitment timing matters means the relevant decisions often fall years before succession planning engages the next generation directly. Second, for internationally mobile families, the relevant labour market may be one the family's advisers do not operate in. Third, financial succession and professional development are distinct: a plan to transfer wealth responsibly says nothing about whether a child can pass a case interview in London or a Super Day in New York.

We do not assume that private banks should provide careers advice. Doing so would raise obvious questions of competence, scope and conflicts of interest. The narrower question is whether trusted advisers recognise the issue, where they currently refer families, and whether those referrals are evidence-led.

Evidence status. Survey figures and programme descriptions are published (fact). The absence of research on adviser practice reflects our search, not proof that none exists (observation). Whether a gap exists in practice is a hypothesis for the interview programme in the section on what we still do not know.

How AI may change the transition

Three developments need to be kept apart.

Candidate use of AI. Students are using AI in job applications and, in a minority of cases, in assessments. Prospects' Early Careers Survey 2025 found that one in ten students had used AI to complete online tests or during interviews [Prospects Luminate 2025]. ISE reported that 85 per cent of its employer members had encountered some form of cheating in their application and assessment processes [ISE 2025].

Employer rules on candidate use. The pattern in the Who Gets In? tracker is a boundary rather than a ban: help with written materials is generally permitted, while timed tests and interviews are protected (see access, inequality and information). Some employers now build AI into an assessment by design, but confined to named stages or roles [WGI AI Rules Tracker v2.1].

Employer use of AI. 11 of the 37 employers with a statement disclose using AI somewhere in their own selection process, from an AI avatar presenting prerecorded interview questions to AI-assisted matching, while 16 publish a statement that people make or check hiring decisions [WGI AI Rules Tracker v2.1]. KPMG UK, for example, states that its AI-generated interviewer presents the questions but that recorded answers are reviewed and scored by its Student Recruitment team; McKinsey states that no candidate is rejected solely on its chatbot's output [WGI AI Rules Tracker v2.1].

Candidates appear to believe something different. In Prospects' 2026 survey of about 5,000 students and graduates, 95 per cent believed employers use AI to shortlist applicants [ISE and Prospects 2026]. The two figures measure different things in different populations and cannot be compared as rates. But the contrast suggests a perception gap: many candidates may be preparing for an automated filter that, on employers' own published accounts, is less common than they assume, while under-preparing for the human, timed and spoken stages that employers explicitly protect.

The wider labour market evidence is early. Brynjolfsson, Chandar and Chen, using US payroll data, found that workers aged 22 to 25 in the most AI-exposed occupations experienced a 13 per cent relative decline in employment after the spread of generative AI, controlling for firm-level shocks [Brynjolfsson et al. 2025]. ISE's members reported graduate hiring down 8 per cent in 2024/25 and up 4 per cent in 2025/26 [ISE 2025; Wonkhe 2026]. Neither establishes what AI is doing to entry into banking, law, private capital or consulting specifically.

Evidence status. Survey and tracker figures are published with stated samples (fact). The perception gap is our interpretation of two non-comparable figures (inference). Effects of AI on entry to the four sectors are not established.

A framework for navigating education-to-career decisions

The evidence above suggests that "elite education leads to elite careers" compresses five distinct stages. Exhibit 1 separates them. It is an analytical model for organising evidence and questions, not an empirically validated instrument.

Exhibit 1. The education-to-selection gap: an analytical model

The model's main implication is that most families' investment concentrates at Stage 1, while the evidence base for Stages 2 and 4 is thinnest. That does not mean families should spend less on education. It means the stages are distinct, and that strength at one does not substitute for attention to the others.

Exhibit 2. Cross-sector selection comparison (published evidence, illustrative not exhaustive)

Sources: WGI register signals cited in Sections 3 and 4; Yale OCS 2026; Apax 2020 via WGI SIG-316, SIG-317. Private capital rows rest on thin public evidence and are labelled accordingly.

Exhibit 3. Who advises families, and where

No adviser in the table is positioned by default to cover Stages 2 to 4 for a specific family across several markets. Whether that matters depends on the family, the child and the route. Many candidates succeed through university careers services, employer programmes and their own research, at no cost.

Implications for families and their advisers

These are recommendations from the evidence, not prescriptions, and none requires private advice.

For families. Treat the education-to-career transition as a separate project from university admission, beginning in the first year of university or earlier for US routes. Read each target employer's own process pages, not secondary summaries. Note application limits and deadlines that fall before the published closing date. Ask the student what each stage tests, rather than whether the grades are good enough.

For universities. Publish sector calendars to first-year students and their families, including market-specific differences for international students. Close the awareness gap identified in survey data on careers service use.

For education consultants. Be explicit with families about where admissions expertise ends. A strong university outcome is a starting position for a second contest with different rules.

For private wealth advisers. Recognise that next-generation conversations about purpose and responsibility may sit alongside a professional ambition with fixed and early deadlines. Where referrals are made, ask what evidence the recipient works from.

For employers. The silence of many leading firms on AI rules, and the gap between candidate perception and employer statements, are both addressable by publication. Clear public rules reduce the advantage of private information.

For candidates. Prepare for the stages employers say they protect: timed reasoning, spoken explanation and case judgement, done unaided.

What we still do not know

The evidence leaves important questions open. We set them out as a research agenda.

1. What do families with professional ambitions for their children know about specific selection processes, and when do they learn it? No survey we located addresses this.
2. How often do clients raise a child's career with private bankers, family offices or private-client lawyers, and what happens next?
3. Do published academic screens match practice, and how far do contextual recruitment policies change outcomes?
4. Which forms of preparation improve performance at Stage 4, and are their benefits equally distributed?
5. How do private capital firms weigh a candidate's first employer, and does that vary by market?
6. Is the timing of recruitment moving earlier or later across markets, measured against the same programmes over several cycles?
7. What, if anything, is AI changing in entry to these four sectors specifically, as distinct from the wider graduate market?

Who Gets In? intends to address questions 1 and 2 through structured interviews with private bankers, family-office advisers, education consultants, university careers leaders and employers. Interview findings will be published separately from this desk research, with a stated methodology and consent arrangements.

Conclusion

Elite education matters. It is strongly associated with access to the most selective careers, and in the best US evidence it causally raises the chance of working at a prestigious firm. But the evidence does not support treating education as the whole of that access. Employers run their own contests, with their own screens, formats, calendars and rules, many of them published and many not. The decisive applications often fall earlier than a family focused on university outcomes would expect, and the capabilities tested differ from those that university assessment rewards. The strongest gaps in the evidence lie in the middle of the transition: what families know, when they learn it, and what preparation works. Those are answerable questions, and the people best placed to help answer them include the advisers families already trust.

Research methodology

Scope. Entry into investment banking, private equity and private capital, corporate law and strategy consulting, with priority markets of the United Kingdom, United States, Singapore, Hong Kong and the UAE and GCC. The article is desk research. No interviews or surveys were conducted for it.

Sources. Three classes: (1) the Who Gets In? register of official employer and university sources, read on 9 November 2026 (1,387 signals from 494 official source records, last canonical update 2 November 2026), and the AI Rules Tracker v2.1 (67 employers, observed 23 October 2026); (2) peer-reviewed research and established research organisations; (3) employer and professional body surveys with disclosed samples. Secondary reporting is used only where a primary source was unavailable, and is marked.

Who Gets In? method. Each signal records one observation from one official source for one institution, programme, market and date. Samples are purposive, not representative. Absence of a public statement is recorded as unknown, not as absence of practice. Published processes are not evidence of practice or enforcement.

Evidence classification. Claims are marked as fact (published data or statements), observation (what we found or did not find), inference (our reasoning from facts), hypothesis (a proposition to test) or recommendation.

Limitations. The register is weighted towards London, banking and law; private capital and GCC processes are thinly documented. Employer statements describe policy, not practice. Survey figures come from different populations and years and are not combined. US sources dominate the causal literature; UK and Asian equivalents are scarcer. Dates of observation are given for every register citation.

Dates. Register and tracker data as stated above. External sources accessed 9 November 2026.

Source-verification register

Status: V verified on the source page; V (secondary) verified as reported by a named secondary source, original not opened; U unverified. Checked 9 November 2026.

Institutional disclosure

Who Gets In? is a research programme run by Elite Careers Strategy, an international private advisory supporting individuals and families pursuing the world's most selective careers. Elite Careers Strategy is a paid advisory firm working in the field this research examines. This article uses public sources only and contains no client evidence. The research is led by Hassan Akram, Founder and Principal Advisor of Elite Careers Strategy. No affiliation with, or endorsement by, any institution named is implied.

References

• Ashley, L., Duberley, J., Sommerlad, H. and Scholarios, D. (2015). A qualitative evaluation of non-educational barriers to the elite professions. Social Mobility and Child Poverty Commission. assets.publishing.service.gov.uk
• Bridge Group (2020). Who gets ahead and how? Socio-economic background and career progression in financial services. As reported in Sutton Trust (2021), APPG on Social Mobility presentation pack. www.suttontrust.com
• Brynjolfsson, E., Chandar, B. and Chen, R. (2025). Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence. Stanford Digital Economy Lab working paper. siepr.stanford.edu
• Careers & Enterprise Company / Gatsby Foundation. Talking Futures, citing CEC Parental Engagement: Evidence Review (2021). www.gatsby.org.uk
• Chetty, R., Deming, D. J. and Friedman, J. N. (2026). Diversifying Society's Leaders? The Determinants and Causal Effects of Admission to Highly Selective Private Colleges. Quarterly Journal of Economics 141(1), 51 to 145. Working paper: www.nber.org ; summary: opportunityinsights.org
• Citi Private Bank. Next generation programme. privatebank.citibank.com
• Daily Pennsylvanian (2018). Goldman Sachs and JPMorgan will no longer recruit sophomores. www.thedp.com
• Dale, S. B. and Krueger, A. B. (2014). Estimating the Effects of College Characteristics over the Career Using Administrative Earnings Data. Journal of Human Resources 49(2), 323 to 358. www.nber.org
• Independent Schools Council (2025). Press release, 15 May 2025. www.isc.co.uk
• Institute of Student Employers (2025). Student Recruitment Survey 2025. ise.org.uk
• Institute of Student Employers (2025b). ISE top 10 stats of 2025. ise.org.uk
• ISE and Prospects at Jisc (2026). Students use AI to compete, yet oppose AI assessment. ise.org.uk
• Laurison, D. and Friedman, S. (2016). The Class Pay Gap in Higher Professional and Managerial Occupations. American Sociological Review 81(4). works.swarthmore.edu
• NACE (2026). Intern conversion rate hits highest mark in five years. naceweb.org
• Private Banker International. Next-generation engagement by private banks. www.privatebankerinternational.com
• Prospects Luminate (2025). Early careers recruitment in 2025: integrity, inclusion, entry standards. luminate.prospects.ac.uk
• Prospects Luminate (2026). Careers advice and confidence: insights from A-level students. luminate.prospects.ac.uk
• Rivera, L. A. (2012). Hiring as Cultural Matching: The Case of Elite Professional Service Firms. American Sociological Review 77(6), 999 to 1022. doi.org
• Spear's (2026). HSBC and MIT Sloan launch programme for UHNW heirs. spearswms.com
• Sutton Trust and Social Mobility Commission (2019). Elitist Britain 2019. www.gov.uk
• Sutton Trust (2026). Private Tutoring 2026. www.suttontrust.com
• Times Higher Education (2026). Employers cut student work experience roles as demand rises, 14 October 2026. timeshighereducation.com
• UBS (2025). Global Family Office Report 2025, press release, 21 May 2025. www.ubs.com
• Who Gets In? (2026). Signal register, read 9 November 2026. whogetsin.com (individual signals at whogetsin.com)
• Who Gets In? (2026). AI Rules Tracker v2.1, observed 23 October 2026. whogetsin.com
• Wonkhe (2026). ISE's 2026 student recruitment survey. wonkhe.com
• Yale Office of Career Strategy (2026). Investment Banking: 2028 Summer Internship Overview (For Sophomores). ocs.yale.edu