The National Education Policy published recently, has generated much activity to re-imagine Indian Higher Education. Since many of the recommendations there appear to be inspired by models in the United States, it may be interesting to report on one small but crucial aspect from  experience is from a leading US technological institution. Namdeo[1] for instance, has written about the idea of Internships that is actually quite active in India, as well as the USA.

H1B From The Other Side

At the turn of the Millennium, US industry demanded a large increase in H1B visas, citing the “Y2K” fear and overall shortage of STEM-trained professionals. STEM stands for Science, Technology, Engineering and Mathematics. US Congress (COTUS), in approving the increase, specified that 1/3 of H1B application fees be given to the National Science Foundation (NSF), to increase domestic supply of STEM graduates. NSF’s Division of Undergraduate Education initiated the “CSEMS” programme: Computer Science, Engineering and Mathematics Scholarships. Each institution could propose one programme of up to $100K/year for 4 years, money to be disbursed to undergraduates based on Merit and Need. A full-time student could receive the lower of $3300/yr, or their Established Need under FAFSA (Federal assessment of financial need). The expected outcome was graduation to STEM careers. Chosen disciplines were limited to CSEM.

The “Deal”

The institution could only take up to 10% of the amount as direct administrative costs, and zero Overhead.  For comparison, a typical university overhead is 60% of direct cost. The PI/PD (principal investigator/director) had to be full-time academic faculty, which ruled out Administrative offices.  The 10% limit dissuaded non-serious proposers.

I was allowed by our Vice-Provost (Academic) to initiate and submit a proposal from our public Type 4 (PhD-granting) university with over 10,000 undergrads spanning the technology spectrum. Some 75% of our funding came from externally sponsored research. So did much of faculty salaries.  However, I knew my colleagues. Our core team had 3 AE and1 ECE faculty, plus the Director of Financial Aid Office who eagerly joined us to boost badly-needed resources.

We boldly proposed to (a) charge 0% “administrative”, leaving full $100K/year as direct scholarships, and (b) enlist several more colleagues as “Mentors” pro bono for the scholars. Both gambles worked well: at its height, over 40 of our most productive faculty across disciplines had readily accepted my “deal”: 0 pay or recognition, hard work, and the chance to mentor some excellent students. We won Phase 1, $400K, 2000-’04.


Students from needy backgrounds do not have an easy time meeting fees and living expenses in US colleges. We were proud that our Institute (then) had one of the lowest fee structures in the nation. A competitive advantage was our ability to recruit students from faraway States because our “out-of-state” fees were less than their “in-state-resident” fees. Many came from families where only one parent was working, jobs in peril in a recession. Some were first-generation college entrants. Many came with feelings of inadequacy among wealthier urban peers brought up by “tech-savvy” parents. Most had to work part-time jobs for instance in grocery stores, to make ends meet. One young lady had come from the Army where she led a Patriot Surface to Air Missile battery team. She was working 40 hours a week in the evening and night at a Kroger supermarket while taking a full load of aerospace engineering classes. One of our top students, she joined my research group with a research assistantship through her Masters Degree. Another was the son of a single mother whose business had gone broke: per the FAFSA their family was well off since they owned a business, so he did not qualify for Financial Aid. He told me of staring at just a loaf of bread, aware that it had to last him a week. He too joined my research group and went on to a PhD at M.I.T. Hunger was a harsh fact of life: you see why those 40 dedicated mentors leaped at my “deal”.

Program Model

Our unique model was driven by experience and tempered by reality:

  1. Scholars were selected  based on the interest (“fire in the belly”) that we saw in their applications including an essay on ambitions, past indicators, SAT scores and grades to-date, if any. Since all applicants had high merit, need probably weighed more. Non-merit criteria were specifically ignored. We hypothesised that with prevailing demographics of need our cohort would beat the Institute’s average in including minorities and women. We proved right – no joy there.
  2. Once selected, a 3.0 Grade Point Average was nominally expected for continuation, similar to our State’s “HOPE” scholarship. However, we specified a sliding scale. Disadvantaged students often suffered from “College Shock”. It was often their first exposure to demand for intense effort, not just talent. The double trauma of losing their scholarship on top of the shock on seeing their grades, and seeing the uphill climb needed to bring GPA back over 3.0, often drove students out of the institution. So we specified that following a “bad” semester, scholars had to show improvement, rather than absolute GPAs, for continuation. Mentoring was a big part of our strategy. We projected that our scholars’ GPA would well exceed the average by graduation. This too proved right, a source of pride.
  3. We had active research programmes into which we would have liked to bring these students (see 2 examples above). However NSF’s Undergraduate Division, endowed with a strong prejudice against “research professors”, specifically prohibited that. My pre-proposal query brought an angry response that they were trying to save students from having to work so that they could focus on studies, not get “cheap labour” for my research labs. We dealt with this as seen below.
  4. Each scholar signed a condition of acceptance, to show up at least twice a semester for discussions with an assigned Mentor. This sounds trivial, but posed a big hurdle for many students, given their feelings of inadequacy that manifested as an anti-authority attitude until they learned better.
  5. Most riskily, we specified that each scholar had to attend at least TWO research seminars given on campus. We quickly realized that scheduling special “FAST Seminars” was a lost cause given the widely differing class schedules across different schools and classes. Instead, students had to find seminars on their own: not difficult as dozens occurred every week across campus. Students were encouraged to find seminars outside their “specialty” schools  – a ploy that worked superbly as they found sparks of inspiration where they would never have ventured.
  6. For only first-year students, one of 2 seminars in a semester could be on Interviewing Techniques, given by the Career Placement Office. This made a big difference to many.
  7. Seminar attendance was proved by submitting a ½ to 1-page summary of what the student learned. This was to be in their own words (the Official Seminar Abstract was also required), with a specified format. They were strongly encouraged to talk to the presenter to gain insights. This was an exercise in listening, learning to grasp the gist of a lecture outside their classes or level, talking to an outside professional, and writing to a specified format. Many students had to go through several iterations before our colleague who read every submission would approve them.
  8. Mentors’ permission was required (an email) to “count” a seminar: Public relations/glitzy presentations from government agencies, companies et cetera, were specifically prohibited. Internal research seminars such as PhD Defences (and even a lecture by our School Chair!) were not counted. It had to be a research seminar delivered by an external researcher – precisely what few undergraduates would consider attending voluntarily!
  9. Mentors opened doors to opportunities such as Internships. Our formal Co-Op programme office coordinated work opportunities with corporations and government labs, where students alternate Work and Study semesters. Seminars opened minds and opportunities for recruitment to all such positions.
  10. A student who won an Internship or Co-Op arrangement was ineligible for scholarship funding that semester, but could resume when back on campus. We counted these as Successes. Any student could sign up for Research Special Problems for academic credit, leading in a following semester to one of our “President’s Undergraduate Research Awards” (PURA) – another “success” that freed funds partially to help another student. These avenues for research participation led many such students to graduate school and careers. Thus we obeyed the NSF Undergrad Division’s prejudice against “research” without letting it hinder students’ opportunities – or our ability to recruit these students into research!


  1. The last few items caused much stress, but our instructions to Mentors were strict. People had to learn to keep promises – or be dropped after warnings. This occurred in about 5 to 10 cases.
  2. In year 2.5 (2002), NSF called. COTUS would not extend CSEMS funding, the recession cutting H1B visas. So NSF kindly asked me to submit a renewal proposal far ahead of the usual timeline, and funded it immediately for another 4 years with available funds. Performance did bring some recognition!
  3. We stretched 8 years’ worth of funding far beyond the 60 students that it should have funded at $3300 per student per year. The Internships, Co-Ops and research lab experiences were multipliers. We took over 200 students to STEM graduation. The vast majority went directly into STEM careers or graduate school: we did not count MBA or careers outside STEM.
  4. Many of those careers were well outside the chosen “specialty field” of the student’s school  – a research seminar, Mentor, lab or internship/Co-Op job had opened eyes and horizons.

All Good Things Come to An End

NSF complimented us on our success. In the next iteration, with a different programme manager, they bowed to complaints from non-performing institutions. They hiked the maximum per student to $10,000/yr, and allowed a significantly larger “administrative cost”: you can guess the cut in number of students who could be helped or induced to seek external support through Internships and Co-Ops. This no doubt brought hordes of proposals, most directly from administrative offices rather than from research-productive faculty. I have not checked whether the programme still exists or thrived. 

Summary of Features for Possible Emulation

Some suggestions are given in Table 1. More details are in the 4 publications listed below.

Research seminars show how technical content is applied; and demand learning to absorb the gist of complex presentations.Required finding seminars, communicating with researchers, and writing clear summaries. Severe resistance until value became clear.
Mentorship as part of scholarships brings some good aspects of “Gurukula”.Many with an “entitlement” view were inclined to just “take the money and run”. A “scholarship” is a valuable “hook” to convey useful traits and resources.
Research-productive faculty should run the program, collaborating with Financial Aid and Career Counselling professionals.Able to tackle urgent financial problems and present a unified “front” to induce compliance with official authority.
Pressure to eschew clear writing in favor of Elevator Pitches, Podcasts, Presentations, Social Media must be rejected.Modern students are adept with skills other than those of the reflective reading and writing, essential for success in technology.
Review and editing of student submissions must be swift and thoughtful.Timeliness is crucial to fairness and credibility.
Self-help must be taught and emphasized to build well-founded confidence early.The student’s own efforts become a force multiplier for education.
Much learning occurs outside classroom lectures. The good workplace is nearly always “cross-disciplinary”.We learned a great deal through the eyes of the students, about subject areas far outside our own.
Table 1: Lessons learned

I hope that these experiences with our model will benefit teachers in India. More details can be found in the 4 publications listed below.


  1. Namdeo, S., Towards an Internship Policy in STEM. The Science Policy Forum., August 24, 2020.
  2. Komerath, N.M., et al., Mentoring Students To Technology Careers. Paper 2004-1228 Division: Minorities in Engineering, Proceedings of the ASEE Annual Conference, Salt Lake City, UT, June 2004.
  3. Komerath, N., McTier, et al. Partnership for Mentoring: The Georgia Tech. CSEMS Program At Age Six. Proceedings of the ASEE Annual Conference, Pittsburgh, PA, June 2008.

DISCLAIMER: The author is solely responsible for the views expressed in this article. The author carries the responsibility for citing and/or licensing of images utilized within the text.