Application Review Request: Exceptional Promise - ML Engineer (1.5 YOE) [OC3 & OC4]

Stage 1
1

Hello everyone,

I am preparing my Stage 1 endorsement application for the Global Talent Visa (Exceptional
Promise) as a Machine Learning Engineer with about 1.5 years of industry experience.

Recommendation Letters

  • Rec Letter 1 (Academic Focus): Written by my University Research Supervisor (UK). Focuses on my academic, the novelty of my Medical Imaging AI research.
  • Rec Letter 2 (Technical Scale Focus): Written by the VP at my current product-led
    digital technology company. Focuses on my technical ownership, highlighting that despite my
    early career stage, I architect and manage deterministic ML infrastructure.
  • Rec Letter 3 (Commercial Vision Focus): Written by a former Manager who is also the CEO of the Company. Focuses on my intrapreneurial spirit, explaining how I bridged Data Science
    research to board-level business strategy, driving revenue through tech adoption.

Mandatory Criterion (Emerging Leader)

  • Evidence 1: AI Collaboration and Entrepreneurial Award in an incubator setup (University led 2 months programme in 2023)
    A formal letter on letterhead. + Award Picture.
    Confirms I won an entrepreneurial award for “Collaboration in AI.”

  • Evidence 2: Commercial Impact & Board Adoption Case Study. (2024-25)
    A project that required a Client to gain Confidence in the Recommendation System and Derived Revenue Signed off by the Client CEO.

Optional Criterion 4 (Academic Excellence) (2023)
Focus: Research published or endorsed by an expert.

  • Evidence 3: International Conference Presentation Proof.
    Combined PDF of my accepted paper abstract and the official conference program - ICMLMI. Highlights that Medical Imaging AI research was accepted for an “Oral Presentation” at
    recognised international conference of Machine Learning for Medical Imaging.

  • Evidence 4: Dissertation Nomination Letter. (2023)
    Factual confirmation letter from my University Course Admin. Officially confirms my dissertation was nominated for the “Best Dissertation Project Award” for the year 2023/24 - amongst Top 3 for the Year in University (UK).

Optional Criterion 3 (Product Led Technical Contribution) (2025)
Impact, scale, and technical complexity at a product-led tech company.

  • Evidence 5: Technical Architecture & Complexity Report.
    High-level architecture diagrams of the High-Dimensionality Data Pipelines and systems I designed. Includes screenshots of my Git commit history to prove my individual contribution. 3-page document signed by a Lead Data Scientist.

  • Evidence 6: Scale & Impact Dashboard. (2025-26)
    Verified system metrics signed by an Engineering Manager.
    Same Architecture based project showing the Outcome Metrics- Dashboard showing System scaled from 10 to 150+ active clients.

Medical Imaging Project got multiple Recognitions - Evidence 3, 4.
Can I use this as two evidences, with different depth and Scope tailored for their respective requirements?

Would really Appreciate your Feedback on the evidences and any Gaps.

Thank you in advance for your time and feedback!

2

Hi @shubham, hope you’re doing well.

First, regarding your recommendation letters. The letter from your University Research Supervisor might be a bit tricky. Academics aren’t automatically considered sector experts, so you’ll need to confirm that your supervisor has a strong profile that clearly shows expertise beyond academia. For the VP and CEO, they can absolutely work, but make sure their LinkedIn profiles clearly reflect seniority, industry impact, and a solid career trajectory. Their credibility matters just as much as what they write about you.

On the award for “Collaboration in AI” awards can be okay, especially if they’re within the last five years and show recognition of your potential in the tech sector. But right now, the way it’s presented sounds a bit vague. You’ll need to clearly explain what you actually did. What problem did you solve? What was the measurable impact? The award itself isn’t enough, the assessors want to see how your demonstrates emerging leadership in tech.

For the client who gained confidence and signed off revenue, on its own, that doesn’t necessarily show sector recognition and the question is were you recognised for that work? If there’s formal acknowledgement, testimonials, or executive validation, that will strengthen it. But still not strong for a second MC evidence.

OC4
The conference presentation at ICMLMI can be a strong piece of evidence, but only if you clearly show the conference is reputable and your paper and presentation topic are relevant to the field. Don’t assume they assessor should know. And also acceptance alone is not enough. Did you actually present? Was the paper published in proceedings or a journal? Has it been cited? Can you provide proof of publication? Even better can you get a senior academic (ideally not your supervisor) to write a short letter confirming the quality and impact of your research work? That would significantly strengthen this evidence set.

Regarding the dissertation nomination, it helps your narrative, but OC4 is about demonstrating contribution not just recognition. You need to clearly show how your research advanced the field. Also, keep in mind that research done purely as part of an undergraduate or MSc thesis usually does not qualify under this criterion. If that nomination led to something further grants, scholarships, further research funding, collaborations that would make it much stronger.

For OC3
Your architecture diagrams and Git commit history are good technical proof if they show ownership and are substantial. But I’d strongly suggest getting a reference letter from a senior executive who can clearly explain how your technical decisions led to commercial improvement. The link between technical delivery and business impact needs to be explicit.

The dashboard showing scale from 10 to 150+ clients is helpful, but if it’s not externally validated, it could be seen as self authored. Having it on an external analytics platform like Google or clarity can help. Also, you are technical and have little or nothing to do with marketing, so you want to be clear about what you did technically that translated into user growth and adoption.

To answer your question about using the Medical Imaging project as two separate evidences, yes this can work, as long as each piece serves a different purpose and clearly meets the specific requirement of the criterion it’s being used for. Just make sure you’re not repeating the same story without adding new depth or scope.

Overall, your conference presentation and the technical architecture contribution are probably your strongest pieces right now. I would recommend working on the other areas, especially external validation, and clearer proof of impact to improve your chances of endorsement.

All the best.