JetBrains Research Podcast

Could human-made ideas become a luxury in a world of AI-generated content? Join machine learning researcher and educator Anastasia Yanina, Igor Slinko, and Ilya Zakharov as they explore the future of ML research. Drawing on Anastasia's experience at Samsung AI Center and Meta Reality Labs, they discuss what foundation models leave for researchers to invent, AI-written papers and peer review, and the value of human creativity. The conversation also covers PhDs, teaching, first jobs, AI coding agents, and whether human code review will survive.

Chapters

00:00:00 Teaser

00:00:27 Welcome and introductions

00:01:25 From art and architecture to machine learning

00:08:13 Switching between computer vision and LLMs

00:10:35 Building 3D avatars at Meta Reality Labs

00:14:01 What's left for ML researchers to invent?

00:19:35 When AI writes and reviews research papers

00:22:39 Scientific publishing and the peer review crisis

00:30:25 Can AI agents reproduce research papers?

00:36:34 Bringing research into industry

00:40:06 Will human-made ideas become a luxury?

00:43:11 Is a PhD still worth it?

00:48:24 Teaching and assessment in the age of AI

00:55:01 Finding your first job in ML

00:59:39 Should students still learn to code?

01:02:36 Writing code with AI agents

01:06:30 Reviewing AI-generated code

01:09:03 The future of pull requests

01:16:45 Changing your mind as a researcher

01:20:35 Advice for a new ML researcher

Mentions & Links:

- Anastasia Yanina — LinkedIn: https://www.linkedin.com/in/anastasia-ianina/

-Anastasia Yanina — profile (in Russian): https://www.setters.media/setters-a-list-2025/yanina

- Yandex Machine Learning Handbook (in Russian): https://education.yandex.ru/handbook/ml

- Hugging Face — What We Learned by Reproducing 2,200 Papers from ICML: https://huggingface.co/blog/icml-2026-open-reproductions

- ICML — On Violations of LLM Review Policies: https://blog.icml.cc/2026/03/18/on-violations-of-llm-review-policies/

- Harbour.Space University: https://harbour.space/

In our fifth episode, we chat with Dr. Cat Hicks, Founder and Chief Scientist of Catharsis Consulting. She is a psychologist and author of The Psychology of Software Teams, published July 2026.

Episode links:

• Dr. Hicks' LinkedIn: https://www.linkedin.com/in/drcathicks/
• Catharsis Consulting: https://catharsisinsight.com/
• "No Silver Bullets" paper: https://arxiv.org/pdf/2503.05040

Our previous episodes can be found here: https://jb.gg/rsrch-pdcst

Find out more about JetBrains Research here: https://jb.gg/rsrch

Chapters:

00:00 Intro

00:46 Meet Dr. Cat Hicks, psychologist for humans in technology

04:15 Why write a book about software teams?

08:49 Building a "third place" between academia and industry

17:10 Why publish research when technology moves so quickly?

27:07 Why study teams instead of individual developers?

31:13 Team culture, developer identity, and AI skill threat

37:55 "More bands, fewer rock stars": The lone-genius myth

43:37 Code-review anxiety and code review as a learning practice

47:27 How AI creates uncertainty about belonging and expertise

53:03 Overproduction pressure: When AI becomes a productivity hammer

55:47 Context engineering and scaling human mental models

58:40 Culture is infrastructure—and invisible work matters

1:01:36 Being a public expert when the answers keep changing

1:07:02 What should we research about AI next?

1:11:14 Take-home message: Your actions matter

Join us as Thomas Petricek, a program and language researcher, discusses the diverse "cultures of programming," highlighting how a humanistic approach can foster better understanding and interaction with software, moving beyond mere automation.

• Our previous episodes can be found here: https://jb.gg/rsrch-pdcst
• Find out more about JetBrains Research here: https://jb.gg/rsrch

Chapters

00:00:00 Programming Cultures Overview

00:01:41 Thomas's Background 00:03:50 Current Research Focus

00:06:05 Code Poetry Explained

00:09:00 Critical Software Architecture

00:11:51 Cultures of Programming

00:17:27 Mathematical Culture

00:22:39 Hacker Culture

00:25:32 Tacit Hacker Knowledge

00:30:00 Tacit Knowledge Undervalued

00:38:43 Managerial Culture

00:40:20 Engineering Culture

00:44:48 Spec-Driven Development

00:51:48 Humanistic Culture

00:54:40 Programming Field Pluralism

00:56:48 Personal Cultural Sympathies

01:01:21 AI and Future Cultures

01:04:14 Lessons from History

01:06:17 Making Programming Easy

01:09:09 Future Research Directions

01:11:43 Take-Home Message

01:13:20 Advice for JetBrains

Mentions & Links:

• Tomas Petricek https://tomasp.net/
• "Cultures of Programming" book https://www.cambridge.org/core/books/cultures-of-programming/075A2D1DE611EE47807A683147B21691
• Psychology of Programming Interest Group (PPIG) https://www.ppig.org/
• Code Poetry https://en.wikipedia.org/wiki/Code_poetry
• Whitespace Programming Language https://en.wikipedia.org/wiki/Whitespace_(programming_language)
• CompCert https://compcert.org/compcert-C.html -Smalltalk https://en.wikipedia.org/wiki/Smalltalk
• On the Limits of Making Programming Easy https://tomasp.net/academic/papers/limits/

In our newest episode, we chat with Alexander Kulikov, head of the Computer Science and Artificial Intelligence B.Sc. program at Neapolis University Pafos.

Episode links:

• https://www.linkedin.com/in/alexanderskulikov/
• https://www.coursera.org/specializations/data-structures-algorithms https://cogniterra.org/course/697/promo
• https://compeau.cbd.cmu.edu/online-education-projects/programming-for-lovers/
• https://lean-lang.org/ https://www.manim.community/

Our previous episodes can be found here: https://jb.gg/rsrch-pdcst

Find out more about JetBrains Research here: https://jb.gg/rsrch

Chapters:

00:00 – Introduction and STACS Conference

02:00 – AI in Theoretical Computer Science

04:30 – Bachelor Program and Evolving Curriculums

08:30 – Boolean Circuit Synthesis Competitions

12:00 – The "Game of Life" and Modern AI Iterations

16:45 – LLMs Winning IMO Medals 22:00 – Formal Verification and the Lean Language

25:30 – The Chess Analogy: AI Replacing Mathematicians

33:00 – AI Coding Agents and Don Knuth's $2.56 Check

38:30 – Distributed Algorithms and Complexity Theory

41:00 – Large-Scale Online Education and Adding JavaScript

48:45 – The Importance of Algorithmic Thinking

56:00 – Real-World Application: The Google Maps Polygon Problem

1:01:30 – Bioinformatics, SPAdes, and Fundamental vs. Applied Research

1:10:00 – Sabbatical Dreams: P vs. NP and Linear Time Limits

1:18:00 – Creating Mathematical Animations with Python

1:21:30 – AI Interpolation vs. Extrapolation and Joke Generation

1:25:00 – "Birds vs. Frogs": Research Styles in the AI Era

1:30:15 – Surprises, Rapid Changes, and the Future of AI

1:33:30 – Advice for New PhD Students: Follow Your Passion

1:36:30 – Outro and Final Thoughts

In our second episode, we chat with Ibragim Badertdinov from Nebius about SWE-rebench, coding agents, and his path from dentistry and AI.

Our first episode can be found here

Episode links:

• SWE-rebench: https://swe-rebench.com/
• ConTree: https://contree.dev/
• Ibragim: https://x.com/ibragim_bad

Also mentioned:

• https://huggingface.co/datasets/nebius/SWE-bench-extra
• https://scale.com/leaderboard/swe_bench_pro_public
• https://swe-bench-live.github.io/

Find out more about JetBrains Research: https://lp.jetbrains.com/research/software-engineering/

Chapters:

00:00:00 Teaser & Introduction

00:02:41 Ibrahim's Non-Traditional Background: From Dentistry to Tech

00:10:45 The "School 42" Bootcamp Experience

00:13:00 Getting into Machine Learning & Kaggle

00:17:20 The First Internship & Learning from a Great Mentor

00:20:10 Contrasting the Worlds of Medicine and Tech

00:26:56 What Are Coding Agents?

00:33:40 Explaining SWE-bench: How Coding Tasks Are Evaluated

00:37:15 RLVR (Reinforcement Learning with Verifiable Rewards)

00:43:10 The Creation of SWE-Rebench

00:50:00 The Main Challenges with Task Quality and Scale

00:55:00 What's New in SWE-Rebench V2 (and Kotlin support)

00:58:42 Will AI Change Code Review?

01:02:00 contree.dev: A New Checkpointing & Forking Tool for Agents

01:09:35 The Open-Source Community's Reaction to SWE-Rebench

01:15:20 SWE-Bench Pro vs. SWE-Rebench

01:18:20 Why Nebius Invests in Open-Source Research

01:21:00 Predictions for the Future of AI and Programming

01:26:00 Final Advice for Career Switchers

We discuss the legacy of OpenCV, the future of Computer Vision, how AI can transform professional sports through biomechanics, and why real-world technology adoption remains so slow.

Episode links

OpenCV: https://opencv.org/

FitWise AI: https://www.thefitwise.com/

Anna Kogan: linkedin: / anna-kogan-44b24673