Extreme-Scale, Nano & Fundamental Engineering
/Learn at the frontier
AI for Science Intern
A scientific prediction task with a proper held-out split and a classical baseline reported beside the model.
Engagement
Internship
Structured, hands-on mentorship for a fixed term. This internship is on site at Kolkata, with hybrid days available by prior arrangement. It is not a remote position.
Scope of role
Contribute to live projects. Shadow senior practitioners. Build a portfolio that matters.
Please note this is an unpaid internship — the offer is real work, mentorship and experience at the frontier, not a stipend.
Research classification
Computing quantities that cannot be obtained in closed form, from established physics. Reproducible calculations, convergence and error budgets, released code and datasets.
Computational research. Computed from established physics. A computed number is a prediction, and it is only as good as its convergence, its error budget, and whatever data it can be checked against.
Scale range 10^-100 m to 10^100 m
A model that fits is not a model that predicts. Work from this division reports a held-out result and a competent classical or numerical baseline alongside every machine-learning claim, and states the domain a surrogate is valid over. A digital twin is a model of a system, not the system, and is described that way. Scientific integrity is a condition of every role in this department. Assumptions are stated separately from conclusions. Uncertainty is reported. Negative and inconclusive results are written up, not discarded. Speculative work is labelled speculative, including when that makes it less impressive. Fabrication, falsification, and presenting a simulation as a measurement end an engagement here.
Scale bands are a research classification, not a claim of experimental reach. Most of this span cannot be probed by any apparatus that exists: nothing below about 10^-19 m has been measured directly, and anything at 10^26 m or beyond is inferred from observation rather than engineered. Every opportunity states the kind of work it actually is.
01 — The role
Why this role exists at EduRankAI
Department of Extreme-Scale, Nano & Fundamental Engineering, Division D08: Computer Science, AI & Scientific Computing. This division exists to serve all the others. It builds the research software, makes simulations run at the sizes the science needs, develops AI models for scientific problems with honest baselines, and holds the line on reproducibility and code quality across the whole department. This position: A scientific prediction task with a proper held-out split and a classical baseline reported beside the model. Scale range for this work: 10^-100 m to 10^100 m. Research classification: Computational. Computing quantities that cannot be obtained in closed form, from established physics. Evidential standing: Computational research. Computed from established physics. A computed number is a prediction, and it is only as good as its convergence, its error budget, and whatever data it can be checked against. A model that fits is not a model that predicts. Work from this division reports a held-out result and a competent classical or numerical baseline alongside every machine-learning claim, and states the domain a surrogate is valid over. A digital twin is a model of a system, not the system, and is described that way. Scale bands are a research classification, not a claim of experimental reach. Most of this span cannot be probed by any apparatus that exists: nothing below about 10^-19 m has been measured directly, and anything at 10^26 m or beyond is inferred from observation rather than engineered. Every opportunity states the kind of work it actually is. Level 1 (Research and Engineering Intern). One well-defined problem, delivered with a reproducible artefact and a short written report.
02 — The work
What you will own
- 01 Build research software that other divisions depend on, to a standard that survives being handed over.
- 02 Build and maintain simulation platforms, and make them run at the size the science actually needs.
- 03 Optimise scientific workloads with a measured before-and-after, never an asserted one.
- 04 Develop AI models for scientific problems, always reported against a competent baseline.
- 05 Build digital twins, and state plainly what each one does and does not represent.
- 06 Create data pipelines with provenance, so a number can be traced to the run that produced it.
- 07 Hold the reproducibility standard: seeds, environments, and a result that comes back the same.
- 08 Maintain code quality and testing across the department, including in other divisions' repositories.
- 09 Work on one clearly scoped problem for the duration, with a written result at the end rather than a status update.
- 10 Meet a named supervisor weekly and come to that meeting with what did not work as well as what did.
03 — The expertise
What we look for
04 — The bar
Who thrives here
- → Education: Final years of an undergraduate degree, or a postgraduate student, in a relevant discipline.
- → Experience: No professional experience required. Must be able to show at least one completed project with code or a written result somebody else can read.
- → Scope of the role: One well-defined problem, delivered with a reproducible artefact and a short written report.
- → Must have: Fluent in Python, and able to write code somebody else can read, test and extend. Also: Data structures and algorithms to the level of choosing correctly and justifying the choice.
- → Portfolio: at least one piece of work — code, a written result, a thesis chapter, a preprint — that somebody outside your institution can read and assess. It does not need to be published.
- → Available for a full-time internship for the stated duration, working the hours set out under Terms of Engagement.
Terms of Engagement
How this internship is structured
6 days a week, about 40 hours of total engagement — 5 hours a day of project work and departmental responsibilities, plus 1 hour 40 minutes a day of holistic well-being and personal development.
Type
Full-Time
Working days
6 per week
Rest days
1 per week
Total engagement
~40 hrs/week
Project work
5h per day
Well-being
1h 40m per day
Duration
12 weeks
Project work and departmental responsibilities and holistic well-being and personal development together make up the total engagement above — the total is not all task output. Well-being time covers physical fitness, mindfulness, reading, reflective learning, leadership development and community engagement. About 480 hours of total engagement over 12 weeks — 360 hours of project work and 120 hours of well-being and personal development.
Measured by
Weekly mentor review against a published rubric, plus the completion of the recorded hours. Working materially over the commitment is treated the same as working under it.
Where this stands legally
Offered under the applicable AICTE internship framework, where one academic credit corresponds to a minimum of 45 hours of work — which is why these hours are counted and certified. The engagement sits well inside the 9-hour day and 48-hour week ceiling of the applicable state Shops and Establishments Act, and the seventh day is a full rest day.
At a minimum of 45 hours of work per academic credit, this engagement is equivalent to roughly 10.7 credits. Your institution decides what it awards; EduRankAI records the hours and the work.
The full per-level model is published at Working Hours by Level.
05 — Hiring process
What to expect after you apply
- 01
Application review
Every application is read personally within five business days. We respond either way.
- 02
Take-home or live exercise
Role-specific. Time-boxed. Real problems we are actually working on, not invented puzzles.
- 03
Conversations
Deep technical and values conversations with the team you would join. No trick questions. No panel ambushes.
- 04
Offer or honest no
If yes: digital offer letter, signed in-portal, transparent terms. If no: written feedback if you want it.
The standard
What you deliver, and how it is judged
Deliverables
- Software with tests, a recorded environment, and documentation another division can act on.
- Performance work reported as a measured speed-up with the machine and problem size stated.
- Models reported with held-out metrics and a baseline, never with training metrics alone.
- Pipelines whose outputs carry provenance back to the run and the code version.
Evaluation criteria
- Does somebody else's checkout produce the same result.
- Is the baseline present, competent, and honestly reported.
- Is the performance claim measured rather than asserted.
- Would the code survive its author leaving.
Preferred, not required
Nothing in this list is a bar to applying.
Tools you would work in
Reports to: Computer Science, AI & Scientific Computing Lead
Works with: D01, D02, D03, D04, D05, D06, D07, D09, D10, D11, D12, D13, D14, D15
Before you start
What we will collect. What it costs. What we will not do with it.
We will collect
- Name, email, phone — Account + application updates. No marketing.
- Resume / portfolio link — Human review of your work.
- Date + place of birth — Identity verification only.
- Your written responses — Selection rubric. Read by humans.
- Government ID (later) — Anti-fraud at offer / interview stage. Not at signup.
We will never
- Sell your data
- Share with third-party recruiters
- Use for advertising
- Train models on it
- Send marketing email
Our situation
EduRankAI is a small, independent organization building long-term capabilities in educational intelligence, advanced AI systems, and research infrastructure. We take no advertiser money, no donations with strings attached, and no investor pressure on hiring decisions. Applying is free, and every application is read by a human — recruitment, technical, academic and leadership teams. It buys us the right to be honest.
Ready to apply?
We read every application personally. If you are the right person for this role — regardless of pedigree, background, or where you are based — you will hear back from us within five business days.
Apply through this page. You will be asked for your education, your experience, and links to work we can actually read — a repository, a write-up, a thesis chapter, a preprint. Send the piece of work you would defend, not the one with the best title. If a result in it turned out to be wrong, say so; we would rather read that than not know. Every application is read by a person. We assess applications on evidence of the work. We do not filter on institution, on age, on gender, on caste, on religion, on disability, or on where you are from. If any part of this process is inaccessible to you, tell us and we will change it for you.
Other roles in Extreme-Scale, Nano & Fundamental Engineering
Explore related openings
Intern Internship