My journey to O-1 2026: from first filing and denial to approval without an RFE

Hi everyone. I want to share my experience preparing for and obtaining an O-1 visa. This is not legal advice, but my personal path, mistakes, conclusions, and what ultimately worked.

My name is Aleksey, I’m 23. I work as a Software Engineer at Cisco on the Webex project, a platform for real-time communications. I’ve lived in Warsaw for over 6 years. I started in IT quite early, so by the time I prepared the case I already had a certain professional background, but I didn’t have a ready understanding of how to properly package it for the O-1.

The idea to open a US visa and possibly move in the future came to me in February 2025. From that moment I started actively studying different visa types, reading sources, analyzing cases, chatting in groups, and working a lot with ChatGPT. There was a lot of information, and without systematization it’s easy to get lost.

Pretty quickly I came across a specialized group, started reading the knowledge base, watching how people describe their cases, which criteria they close and which mistakes they make. In the end I settled on the O-1 because it seemed faster than EB-1 and more realistic at the time. Also I considered O-1 as a possible intermediate step to EB-1 in the future.

How it all started

My path began with studying all possible criteria that can be claimed in O-1. Every day I read the chat, looked at what people write about their cases, asked ChatGPT about each criterion, and collected information from different sources.

I also signed up for free consultations with lawyers. My goal wasn’t so much to hire someone immediately as to understand how they view cases, which criteria they most often use, what they consider strong and what weak.

The main goal at the first stage was simple: to understand the logic of O-1 as much as possible and figure out how to build a working case from my experience.

One important conclusion right away: if you need to prepare a case quickly, start with research publications. This is one of the longest criteria, because writing a paper, choosing a journal, submitting, peer review and publication take time. It’s better to start that process as early as possible.

Initially I looked at these criteria:

  1. Critical role.
  2. High salary.
  3. Media.
  4. Associations.
  5. Judging.
  6. Research publications.

At the start it seemed to me that I already had critical role, salary and media, but later it turned out that this was not enough. Overall my final case differed a lot from what I imagined at the beginning. And that’s normal. Understanding changes a lot during the preparation.

Associations

One of the main problems was that I didn’t have memberships in professional associations.

Through chats and analysis with ChatGPT I found the association Hackathon Raptors. I studied how people use it in cases, what documents it provides, how membership can be described and why it might fit the criterion. In the end I successfully joined this association, and later it also helped me a lot with the judging criterion.

How I used ChatGPT in this process: I exported my chat history for the last two years, uploaded it into GPT and asked it to analyze all mentions of this association: pros, cons, what evidence people use, which phrasings might work, and what risks exist.

Then I found another association — ECDMA. It’s not directly about my profile, since they focus on e-commerce and marketing, but they have an IT track. That helped link the association to my field. Also there were programmers among the participants, which helped in the description.

Main point about associations: it’s important to show that they accept people not just for paying a fee, but for achievements. In my case both associations stated this in their documents, which made describing the criterion much easier.

Research publications

The next criterion I started developing was research publications.

I initially had none, so I needed to quickly understand where and how to publish. With ChatGPT’s help I analyzed different directions, journals in Ukraine, the EU and international journals. For me it was important to find a balance between journal quality and publication speed.

Now I understand that this needed to be started earlier. Writing a paper takes time. Studying a journal’s requirements takes time. Correspondence with editors and waiting for publication also take time.

To those who think it’s too difficult: after reading a few articles, studying the journal requirements, and discussing unclear points with ChatGPT, the picture becomes much clearer. It’s not as scary as it seems at first. The main thing is to choose a topic related to your professional experience and write not abstractly but around real expertise.

Judging

The third criterion I initially lacked was judging.

Judging came through the associations I mentioned above. In the end I had three judging engagements: two from Hackathon Raptors and one from ECDMA.

In my view, this is one of the easiest criteria if you have the right source of invitation and documents showing you were invited as an expert, not as a random person.

It’s important not just to attach a certificate or a letter, but to explain:

  1. Why you were invited.
  2. Why the judging requires expertise.
  3. Who else participated.
  4. What the level of the event was.
  5. What exactly you evaluated.
  6. Why this is connected to your professional field.

Media

There are several paths with media.

The first option is to publish in various outlets where you can be presented as an expert.

The second option is interviews if a journalist is interested in your experience.

The third option is platforms where you can find journalists and offer them your expertise.

The fourth option is professional connections.

In my case there were several media publications. The largest platform I was published on had a reach of about 6 million. But as the first filing showed, merely having publications does not guarantee that the criterion will be counted. It’s important to correctly show the significance of the media, the relevance of the publication, and that the piece is about you and your expertise.

First filing

I followed the agent-based scheme.

Under the agent scheme I had:

  1. Agent (individual).
  2. Letters of Intent.
  3. Itinerary.
  4. Agent Agreement.

Documents included:

  1. 3 Letters of Intent.
  2. 1 Itinerary.
  3. 1 Agent Agreement.

At that time it seemed to me that the case was ready to file. I gathered the documents and decided to go with a minimal description: clear, short and to the point.

The structure of the first filing looked like this:

  1. Forms I-129 and I-907.
  2. Cover sheet with a list of documents.
  3. Documents for the agent scheme.
  4. Description of criteria.
  5. List of exhibits.
  6. Exhibits.

I claimed these criteria:

  1. Associations.
  2. Judging.
  3. Critical role at Cisco.
  4. High salary.
  5. Media — 3 publications.
  6. Research publications — 2 research publications plus several publications on HackerNoon.

I filed the case expecting it to go through normally. But I received an RFE of 14 pages. No criterion was accepted. The agent scheme was also not accepted.

I prepared a response to the RFE. The initial memo was about 30 pages, and the RFE response ended up being about 90 pages. The whole case was roughly 800 pages, and the RFE response reached a similar volume.

In the end, after the RFE response, only two criteria were counted, but I still received a denial.

Mistakes in the first filing

After the denial I started analyzing what went wrong. The main mistakes were:

  1. The critical role was weakly described — recommendation letters were too vague, without clear numbers, metrics and concrete impact. Also, the letters were mostly from Senior Developers, not from people with higher-level positions or external authority.
  2. Research publications were presented unconvincingly — I attached HackerNoon pieces but didn’t prove that these publications were truly in demand or contributed to the field. Even if the pieces were reposted by the CEO and hit top lists, to the officer that might have looked insufficiently academic or significant.
  3. The salary was hard to defend — I attached an official source on salaries in Poland, but apparently the officer didn’t accept it or didn’t find it convincing enough.
  4. There were too few recommendation letters — there were only 6 letters. For my case that turned out to be insufficient.
  5. The memo was weak in structure — it had gaps. The descriptions didn’t always follow USCIS requirements for each criterion. The text was close to what was needed but didn’t clearly show that I’m an expert in my field.
  6. Document placement was inconvenient — criteria, exhibits and evidence were not organized as clearly as needed.
  7. The Exhibit List was not detailed enough — there was a list, but it didn’t help quickly understand what exactly was in each exhibit and what it proved.
  8. Key parts weren’t highlighted — important fragments weren’t emphasized, and the officer likely doesn’t have a lot of time to read. You need to direct attention to the main points.
  9. There was a lack of comparisons — I didn’t sufficiently compare associations, media, salary, judging and other elements with external sources or analogs.
  10. I included a weak criterion — one of the criteria pulled the case down rather than strengthened it.
  11. There was no proper description of future work in the US — I should have better explained where I would work, how exactly, with whom, in what format, and why it fits O-1.
  12. The Itinerary was too general — it didn’t provide enough specifics.
  13. Letters of Intent were too short — they lacked details: exactly what I would do, why they need me, what projects, timelines, tasks and expected contribution.
  14. The Advisor Opinion Letter was not from someone in the US — that also, in my opinion, weakened the case.

Conclusions after the first filing

After the RFE and denial I made several important conclusions.

  • First: it’s better to write the memo in detail and not be afraid of length. Short and pretty is not always good. For O-1 it’s important not just to attach evidence but to explain to the officer why that evidence satisfies the criterion.
  • Second: for associations you need to follow USCIS logic strictly. You must show not just membership but selective membership, membership requirements, achievements of members, and that the association is genuinely related to your field.
  • Third: for judging it’s useful to have not only documents about the judging itself but also additional letters. For example, a letter explaining why you were invited to judge a hackathon and why you are suitable as an expert.
  • Fourth: recommendation letters, expert letters and review letters for research publications help a lot. External confirmations strengthen almost any criterion.
  • Fifth: you need to compare. For example, show that the media outlet is indeed significant, that the association selects members by achievements, that among participants there are experts, that the judging was at a level requiring qualified specialists.
  • Sixth: it’s better to arrange criteria from strongest to weakest. That way the officer sees the most convincing part of the case first.
  • Seventh: highlight important parts in bold. The officer should be able to quickly see the essence.
  • Eighth: a good Exhibit List is very important. It should not just list documents but explain what each document proves.

The main overall conclusion: in each criterion you need to show not just a fact, but expertise. Not just “I participated,” but “I was chosen as an expert.” Not just “I published an article,” but “this publication demonstrates a contribution to the field.” Not just “I’m a member of an association,” but “I was accepted for achievements, and this membership reflects recognition.”

Second filing

After the denial I began to fully strengthen the case and rewrite the memo.

From the old case I took only the separate descriptions for the criteria that were counted, but I also improved those.

The final structure of the second filing looked like this:

  1. All forms.
  2. A cover page with a clear indication of where each document is located — this wasn’t a full Exhibit List but a quick navigation sheet: which main blocks are in the case and on which page they start.
  3. A separate description of where and how I will work in the US, what companies these are, what tasks and why my expertise is needed.
  4. A separate block for the agent scheme.
  5. All documents for the agent scheme.
  6. The main memo of 90 pages.
  7. A detailed Exhibit List: exhibit number, short description, what the document proves, page number.
  8. The exhibits themselves.
  9. Recommendation letters — 12 in total. Some letters were very detailed, 5 pages each.
  10. A separate description of the previous denial and what was corrected and strengthened.

In the memo I tried to write as clearly as possible:

  1. What exactly the criterion requires.
  2. What evidence I provide.
  3. Why this evidence fits.
  4. How it demonstrates my expertise.
  5. What external confirmations strengthen it.
  6. Why the officer should accept the criterion.

I also highlighted the most important parts in bold so the officer could navigate the text more easily.

As a result, with this case I received approval without an RFE on the 11th day of review under premium processing.

Consulate

Booking an appointment at the consulate was not difficult. The nearest appointment was about two weeks out, and I took the nearest slot.

The interview itself was quick, without Administrative Processing. The visa was approved, and soon I had it in my hands.

My appointment was at 9:00. I left around 9:05. First I stood in line, waited to give fingerprints and confirm documents.

I heard that at window 5 a woman was speaking Polish and English. No one was standing at her window, but before me she had already given two approvals. I said I could speak different languages, and they offered me to go to window 5. I agreed.

We started in Polish. She asked about my work at Cisco, about my critical role, where I worked and how long I’ve lived in Poland.

Then she asked how I planned to go to the US if I don’t speak English. After that I switched to English, and the conversation continued in English.

Questions about the case were such as:

  1. Where I will work in the US.
  2. For which companies.
  3. How many research publications I have.
  4. Whether I am an AI Software Engineer.
  5. Why the documents state Florida if I will work in New York.

There were also questions about how long I’ve lived in Poland and whether my parents live here. The consul asked for my residence card and confirmation that I filed an application for a new card.

Overall the interview was calm and quick.

How I used ChatGPT and Gemini

I actively used both ChatGPT and Gemini. Each tool had its strengths.

Gemini, in my view, handled writing text, deep research and advice on structure better.

ChatGPT was stronger at analysis, checking logic, finding weak spots and a more dry, formal style.

For finding information about media, journals, citations and other evidence I more often preferred ChatGPT.

My process usually looked like this:

  1. First I wrote the text myself by hand.
  2. Then I improved it through GPT.
  3. Then I asked Gemini to rewrite the text more coherently.
  4. After that I returned the text to GPT for analysis of weak spots.
  5. Separately I asked to highlight the main arguments.
  6. I designed the case structure myself and then described each criterion according to that logic.

In essence, I constantly tossed text between different models, but I made the final decisions on structure and logic myself.

Main tips

If I were starting over, I would do several things differently from the beginning.

  • First — start immediately with research publications, because that takes a long time.
  • Second — build the memo in detail from the start, following USCIS logic.
  • Third — pay more attention to recommendation letters. They should be specific, with numbers, projects, impact, results and an explanation of why you are the expert.
  • Fourth — plan the agent scheme in advance: Letters of Intent, Itinerary, Agent Agreement, descriptions of companies, tasks and future work.
  • Fifth — make a very detailed Exhibit List. This is really important because the officer must be able to navigate the case easily.
  • Sixth — don’t add weak criteria just to increase the number. Better fewer but stronger.
  • Seventh — every criterion should be proven not only by a document but also by an explanation. A document alone doesn’t always speak for you. You need to directly show what it confirms.
  • Eighth — wherever possible, show external recognition: letters, expert opinions, reviews, comparisons, metrics, reach, citations, event levels, significance of organizations.

Main thought: you need to show the officer not just that you’re a good specialist. You must show that you are an expert in your field, that your work matters, that you are chosen, invited, published, cited or recognized not by chance but for achievements.

That was my path: first filing, RFE, denial, full rework of the case and then approval without RFE.

Good luck with your preparation. The main thing is not to treat the case as a set of documents. It should be a coherent story where each document works toward the overall conclusion: you meet the O-1 standard.

Official sources: USCIS Form I-129

3 Likes

Aleksey, thanks for sharing your story in such detail.

What do you think about the first rejection — how fair was it and why did it happen? I mean: why did you make the mistakes that you maybe now admit? Was it a cavalier, “we’ll wing it” attitude hoping it would pass? Didn’t know how to gather and analyze what was actually needed for the case? How did you learn that afterwards? Or was it something else?

You write that if you were at square one, you would have changed, among other things:

planned the agency setup in advance

How could you have planned that in advance and why didn’t you do it right away? I’m trying to understand what you actually did wrong in terms of collecting material and planning the first case — why didn’t you plan everything from the start the way you did before the second submission?

Maybe your answer will prompt others who are assembling a case on their own to take a different approach to gathering materials than the one they’re using now.

6 Likes

Respect that you didn’t give up after the first rejection and managed to get the visa)

7 Likes

Thank you so much for such an engaging and detailed story. I have a very similar situation; unfortunately, I haven’t gone all the way through this path yet )

7 Likes

Egor, thanks for the question. I think the problem in the first case was that I believed in it too much, even though there wasn’t a strong case in fact (I think it might have flown in 2023, but not today), and I had some mental limitations and modesty. For example, for the critical role I was somehow afraid to ask colleagues to write more detailed letters; in the second filing I asked them to write more detailed ones with numbers and metrics, and the same applied to other aspects — I was too modest, and I was modest in the memo description too. I’ll add that what worked a couple of years ago, a year or even six months ago may not work today, because USCIS is constantly changing, and you need to try to stay in step with that trend.

Regarding the agency arrangement, at the first filing I had no idea how to structure it best, what the LOI should look like, and how to explain it to the employer so they’d draft and write it correctly — what contract would be appropriate and what the itinerary should be. In the end I did what everyone did, what the lawyers recommended: LOI — one page, itinerary — not too complicated, and I got an RFE for that. So in the second filing the LOI was much more detailed, with all information up to the work plan included, and the itinerary was as detailed as possible with tables spanning 5 pages. So I think you need to lay out in your head how the contract will look, what will be in it, and what’s essential (depending on the field) — what needs to be in the LOI (not just the text “we intend to employ you”) to show the officer how, where and why you’ll work, what you’ll do, how the work will proceed, and everywhere add maximum detail so that someone who doesn’t know you can immediately understand what you’ll be/where you’ll be/how you’ll do it.

In short: follow trends and see what works now and what USCIS expects, don’t be modest, put yourself in the officer’s shoes and think whether a stranger reading this will understand what it’s about and have the full picture, and don’t skimp on documentation — a good description with solid attachments is important. Also pay attention to the case structure so it’s easy to read, easy to move from criterion to criterion, and watch every detail. Make it easy for the officer to grasp the essence immediately, understand everything at once, and without problems.

7 Likes

Modesty in recommendations won’t work; the officer needs numbers and concrete details for each item.

8 Likes

That point about being specific in the letters is spot on — I remember a guy who, on his second filing, got EB1A approved in 8 months. He only changed the wording in the letters from generic statements to actual numbers about impact and percentages for the project. Recommendation letter writers often don’t know what to write, you have to tell them straight what metrics the officer needs )

6 Likes