People in AI

Lucas Gonzalez

Engineer

Anthropic

Lucas Gonzalez spoke about the expanding toolkit for Claude at Code with Claude SF.

Acknowledged for contributions to Anthropic's agent security initiatives.

Ludo Antonov

Engineer

Co-founder at SuperMe

Ludo Antonov, co-founder of SuperMe, previously served as head of engineering at Pinterest, Lyft, and Whatnot and is building AI expert networks.

Byline co-author on Cosmos-H-Dreams surgical world modeling system.

Luke Phillips-Sheard

Engineer

Author at Vercel

Wrote article describing Vercel's Fluid compute layer that dynamically configures infrastructure for different workloads.

Lydia Hallie

Engineer

Engineer at Anthropic

CC engineer Lydia Hallie strongly recommends developers to enable auto-fix feature and slack off from manual PR fixes.

Madeeswaran Kannan is a co-author of the Mellum 2 technical report on the 12B-parameter MoE language model.

Collaborated on TurboQuant research for vector compression.

Co-author on Google Developers Blog post about unified API for AI model routing on Google Cloud API Gateway.

Software engineer at Google Data Cloud exploring evaluation methods for AI agents using information theory.

Manav Shah

Engineer

Raindrop

Manav Shah co-authored the Signals 2.0 launch announcement for Raindrop's binary classification model.

Co-author of the article on building a custom Vision LLM for document processing at Grab

Hostinger AI tech lead describing recent pricing changes as the first real test of frontier model pricing power.

Manthan Gupta

Engineer

AI Research Engineer

Built Auto-Inference-Optimiser to study LLM inference optimization by letting an AI agent hill-climb on inference speed while keeping quality fixed on Apple Silicon.

Contributed a PR adding pricing information for OpenAI's o4-mini-deep-research and o3-deep-research models to llm-prices.com.

Mara Kiefer co-authored the GitHub post on improving token efficiency in agentic workflows.

Heads Adobe's Project Indigo experimental camera app adding AI photo critique and editing suggestions.

A robotics adviser to Colin Angle's Familiar Machines who founded Boston Dynamics, maker of the four-legged Spot robot.

Marcin Plaza

Engineer

Engineer at Midjourney

Engineer at Midjourney and tech YouTuber who produced a behind-the-scenes video tour of the company's medical scanner hardware.

Co-authored technical report on Azure ND GB300 v6 Virtual Machines achieving record 1.1 million tokens/s on Llama2 70B inference.

Mark Jen

Engineer

Software Engineer at Meta

Founding team member at Thinking Machines Lab who was hired by Meta; previously worked at Meta.

Co-author of the article on Meta's Backbone connectivity scaling for AI

Co-chair of IETF AI Preferences working group acknowledged in the robots.txt policy token draft.

Mark Smith

Engineer

Infrastructure Lead at Discord

Built Discord's internal WebRTC voice and video infrastructure, demonstrating full-stack vertical integration approach to infrastructure.

Author and developer of the flyhard project training a fruit fly connectome-based model to operate vehicle controls.

Explained that AI shifted debugging bottleneck rather than eliminating it and that engineers now spend more time debugging than writing code.

Marko Kojic is a co-author of the Mellum 2 technical report describing a 12B-parameter MoE language model optimized for coding and agentic workflows.

Senior Manager and co-author of Spotify article on evaluation funnel approach to LLM evals and A/B testing.

GlobalFoundries engineer contributing industry perspective to the defect detection research

Co-author of draft-ietf-aipref-vocab-08 AI vocabulary and co-author of attachment draft with Gary Illyes on associating AI usage preferences with content.

Co-author of the article announcing new NotebookLM features

Maryam Motamedi

Engineer

Researcher at NVIDIA

Co-author on NVIDIA Nemotron 3.5 Transcribe multilingual speech model with transcription, translation, and speaker attribution capabilities.

Mason Drxy discussed LangChain 1.1's capability introspection and Deep Agents for dynamic model routing.

Prime Intellect engineer who collaborated on RL training experiments and provided hardware resources.

Mathew Hogan

Engineer

Cursor

Mathew Hogan co-authored an article describing how Cursor built development environments for cloud agents to test and merge code.

Mati Roy

Engineer

xAI

xAI employee who disclosed the existence of 'spicy mode' in a now-deleted X post about Grok Imagine's nudity capabilities.

Matilda Ankargren

Engineer

Staff Data Scientist at Spotify

Co-author of Spotify Engineering article on LLM evals and experiments.

Contributed to the analysis of Claude Opus 4.6's eval-aware behavior on BrowseComp.

Matt Colyer

Engineer

Product Leader at Figma

Product leader at Figma building tools for designers and developers; authored article on Figma canvas opening to AI agents.

Matt DesLauriers

Engineer

Developer who created Windfoil, a GPU-friendly algorithm for vector graphics rendering that runs in browsers

Using Fable 5 to write world-model training code when Anthropic flagged it as frontier AI research and steering vector kicked in implementing JEPA.

Matthew Carrigan

Engineer

Hugging Face

Matthew Carrigan is referenced in an X thread about GPT OSS innovations mentioned in the Hugging Face blog post.

Co-authored guide on building production commerce agents with Claude, covering latency, cost, memory, safety, and evaluation practices.

Co-founder of Oumi, former Google AI and Microsoft engineer.

Engineering intern who directed Copilot through the process of onboarding new token types

Maurice Heumann

Engineer

Developer

Used generative AI with seventeen agents working simultaneously to decompile Modern Warfare 2 on PC, achieving playable state before receiving DMCA takedown from Activision.

Stated that Eve defaults to 1M context for plaintiff attorney work, enabling cross-referencing of long documents and case files.

Said 'I probably spend more than my salary on Claude' demonstrating tokenmaxxing behavior among AI-obsessed workers.

Lead author of Anthropic's engineering article on containing Claude agents and security failures during evaluations.

Co-author of Google blog post on AI-powered contrail avoidance trials in Asia-Pacific.

Byline author covering Liquid AI's efficient long-context encoders for CPU deployment.

Proposes pairing SKILL.md files with PITFALLS.md files to help AI coding tools avoid repeating mistakes across sessions.

Acknowledged for contributions to Anthropic's agent containment research.

McLaren Stanley

Engineer

Senior Principal Engineer at Amazon Stores

Leads 70-person team modernizing 17 years of code for Amazon's mobile shopping app using AI agents.

Co-author on NVIDIA Nemotron 3.5 Transcribe multilingual speech model project.

SCE senior engineer discussing the utility's grid monitoring platform using AI algorithms.

Michael Feathers

Engineer

Software Developer at R7K Research & Conveyance

Coined the term 'characterization testing' in his book 'Working Effectively with Legacy Code'

Michaël Gharbi

Engineer

Reve

Author of "Reve Launches AI-Powered Platform for Precise Image Creation" in Reve

Author of comprehensive technical guide explaining how prompt caching works across different LLM providers and optimization strategies.

Michael Isaac, a PhD student at Carnegie Mellon University, implemented Flex for DSPy during his internship at cmpnd and authored the guest post introducing the module.

Author who reverse-engineered Claude's generative UI implementation and built terminal version using Glimpse.

Michael Luo

Engineer

ML Engineer at Together AI

Trained the DeepSWE RL model, developed Kubernetes wrapper for R2E-Gym, and implemented agent/environment abstractions for rLLM.

Lead author of Walrus, a transformer-based foundation model for simulating fluids, liquids, and gases across diverse physical systems.

Lead technologist for EUV light sources, discusses the new 1,000-watt system capability and future potential to reach 1,500-2,000 watts.

Used Fable 5 to one-shot a clean interactive architecture diagram as pure SVG and HTML after the model read an OpenAPI spec.

Thanked for reviewing the code review article

Co-authored the TorchTPU engineering blog post on TPU-PyTorch integration.

Michelle Lim

Engineer

Developer at Stripe

Michelle Lim works at Stripe and conducted an informal benchmark testing how well AI models can build and deploy fully functioning t-shirt stores.

Co-authored research on Multi-Token Prediction optimization for on-device LLM inference on Pixel devices.

Co-author of guide on fine-tuning NVIDIA Cosmos Predict 2.5 with LoRA/DoRA for robot video generation.

Mihai Lis

Engineer

Head of Engineering at AMP PBC

Head of Engineering at AMP who worked at Google building Borg/GQM scheduler, focused on compute optimization and scheduling infrastructure.

Used Claude Code to disable bloat apps, replace stock launcher, and add rotating wallpapers to smart TV.

Co-author of Anthropic engineering article on agent containment and security incidents.

Mike Chambers

Engineer

AWS

Co-author of AWS Velocity Labs release of Strands Decider 2B decision model for agent workflows.

Author of "Google Chrome"

Co-authored a blog post on Google Developers Blog about on-device function calling in Google AI Edge Gallery.

Co-author of Walrus, contributing to the development of a cross-domain foundation model for physical simulations.

Acknowledged for contributions to Anthropic's agent containment research.

Mojdeh Rastgoo-Lemaitre oversaw the project using Microsoft Discovery to assemble a knowledge base from 106,000 patents and abstracts in hours, then develop cement-additive hypotheses in six weeks.

Acknowledged for contributions to Anthropic's agent security work.

Monica Phang

Engineer

Student at Boston Consulting Group

Developed a GPT that builds character sheets for roleplaying games

Co-authored article on Claude Code team's code review process

Speaker at AIE Security track on AI security and financial systems.

Moritz Stephan

Engineer

Chief of Staff at Cognition

Engineer who serves as Scott Wu's chief of staff, joined Cognition after seeing Devin build Conway's Game of Life from scratch as a Stanford undergraduate.

Acknowledged for contributions to building Morphex, an AI-powered migration system that reduced monolith splitting from 8 years to 6 months.

Credited for long-term work on competitive programming

Byline co-author on Cosmos-H-Dreams surgical world modeling system.

Motoi Ishibashi

Engineer

Rhizomatiks

Co-developer of the fencing tracking and visualization system using AR and deep learning.

Built dumbtv using DeepSeek V4.1 Flash, Claude Opus 5.5, and Claude Sonnet 5.5 to remove ads and bloatware from smart TVs.

Nan Jiang

Engineer

Engineer at Modal

Modal engineer providing feedback on RL experiment results and sandbox infrastructure.

Posted criticism that Block pushed AI adoption but saw limited productivity gains, refused retention raise and quit.

Nat McAleese

Engineer

Researcher at Anthropic

Said Fable/Mythos had been transformational to daily coding work, progressing from barely useful to barely writing a line of code since Fable.

Nat Sothanaphan

Engineer

Software Engineer

Software engineer and prolific Erdős problems forum participant who used ChatGPT to formalize mathematical proofs.

Natalie Guevara

Engineer

Blog Editor & Strategist at GitHub

Co-author of article announcing the GitHub Multilingual Repositories Dataset for discovering repositories with non-English content.

Co-author of LiteLLM incident report on compromised v1.107.0 release.

Nate Barbettini

Engineer

Member of Technical Staff at Arcade.dev

Explained how MCP session ID changes improve scalability for AI agent infrastructure deployments.

Nathan Brake

Engineer

Machine Learning Engineer at Mozilla AI

Author of "Run Any LLM with a Single API: Introducing any-llm v1.0" in Mozilla AI Blog

Nathan Drezner

Engineer

Product Manager at LangChain

Co-authored announcement of Managed Deep Agents v0.8 with support for user-owned credentials, memory, and HTTP channels.

Engineer working with Clémentine Fourrier on evaluation and leaderboards at Hugging Face.

Created Stringman, a cable-driven robotic claw system powered by LeRobot-based firmware for household automation.

28-year-old computer science graduate and only male member who serves as the group's link to the outside world

Neal Wu

Engineer

Coder at Thinking Machines Lab

Coder who won three gold medals in a programming Olympiad and was hired by Thinking Machines Lab.

Described how IBM uses AI to give new engineers harder assignments sooner, with juniors handling 70-80% of senior-level tasks.

NVIDIA Distinguished Engineer presenting on AI physics and simulation at SIGGRAPH 2026.

Acknowledged for contributions to building Morphex, an AI-powered migration system that reduced monolith splitting from 8 years to 6 months.

Nicholas Lin

Engineer

Anthropic

Author of "Advancing Claude for Financial Services" in Anthropic

Byline author on Cursor's cloud agent development environment systems.

Nick Mcginnis

Engineer

product and UX design lead at Google

Collaborated on the brand identity definition for NotebookLM.

Works on Codex and insisted on social media that the goblin system prompt issue was not a marketing gimmick.

Informed OpenAI President Greg Brockman about a hack found for OpenAI crawlers to bypass NYT paywall.

Community practitioner from One North who created /cc-figma-component skill for generating Figma components from JSON.

Co-author of the 200+ WebGPU Kernels library for browser-based AI inference.

Nicolò Lucchesi

Engineer

vLLM

Co-author of technical post on push-based KV transfer improvements in vLLM for disaggregated inference.

Byline author of article introducing SWE-1.5, Cognition's frontier-sized coding model.

Nikita Dolgov

Engineer

JetBrains

Nikita Dolgov is listed as Pavlichenko in the byline; they are a co-author of the Mellum 2 technical report.

Widely credited for coining the term 'meat proxy' to describe people who blindly copy-paste AI chatbot output without editing.

Disclosed that Spotify engineers run 5 to 10 Claude sessions simultaneously and 73% of pull requests are AI-assisted.

Created CorridorKey, an open-source neural network for perfect green screen keys.

Nikola Jurkovic

Engineer

Member of Technical Staff at METR

Asked clarifying questions about Sol's role in the post-training process for Luna.

Co-author on NVIDIA Nemotron 3.5 Transcribe multilingual speech model project.

Nishtha Bhatia

Engineer

Product Lead at Google

Google product lead for AI & Glasses who demoed Android XR audio glasses ordering coffee via Gemini at Google I/O 2026.

Co-author on NVIDIA Nemotron 3.5 Transcribe multilingual speech model project.

Former MIT visiting student co-developing dual-arm adaptive therapy robot using diffusion models and force feedback.

Noah MacCallum

Engineer

Engineer at OpenAI

Co-authored the Codex Prompting Guide for OpenAI's developer documentation.

PostgreSQL developer praising OSS Scanner for finding defects and providing nearly usable fixes.

Community practitioner from Rad Collab who created /rad-spacing skill for applying hierarchical spacing with variables.

Contributed significantly to the development of the Mavis Beacon program.

Co-author on NVIDIA Nemotron 3.5 Transcribe multilingual speech model project.

Oleg Murk

Engineer

Member of Technical Staff at OpenAI

Teammate credited for preparation and work on IOI competition

Co-authored research on Google's AI-driven flash flood forecasting system that uses news data to predict urban flash floods up to 24 hours in advance.

Co-author of LiteLLM incident report on compromised v1.107.0 release.

Oliver Moazzezi

Engineer

IT consultant at Microsoft

IT consultant who used AI to self-diagnose Lyme disease after doctors attributed symptoms to anxiety

Developer who created RemoveMacAI command-line tool to remove Apple Intelligence from macOS 27.

Vals AI engineer quoted on environmental footprints of agentic tasks scaling dramatically compared to single-shot queries.

Google author of blog post on Gemma 4 QAT models optimization for mobile and laptop efficiency.

Posted practical guides on building AI agent harnesses from scratch and reducing costs through custom harnesses with routing and verification.

Co-founder and machine learning engineer at Oumi, formerly at Apple, Twitter, and Microsoft.

Extropic researcher who co-authored paper on Z1 probabilistic hardware for neural network inference.

Built a GPT to automate creation of social media posts from thought leadership reports

27-year-old former EzDubs founder tasked with refactoring 98% of Cisco's legacy code with AI in 24 months.

Paige A. Willson brought costume design and technology expertise to the MyoStep project.

Boston Dynamics Spot product manager describing how the robot could improve delivery efficiency by allowing drivers to prepare next delivery while Spot delivers cargo.

Palash Awasthi

Engineer

Head of Growth at Reflex

Co-authored benchmark comparing vision agents versus API agents for the same task, showing vision agents cost 45x more in tokens.

Co-author of the article about GitHub Copilot's impact on security workflows

Co-author of the article announcing Gemini Batch API extensions

Google engineer who claimed on X that Chinese models GLM-5.2 distilled Claude and GPT 5.5.

Contributed to the analysis of Claude Opus 4.6's eval-aware behavior on BrowseComp.

Byline author who wrote about MCP 2026-07-28 spec release and its features for connecting AI agents to applications.

Discussed forward-deployed engineers and software factories at AIEWF session on agent deployment.

Referenced in "AMA with the OpenAI Codex Team" in OpenAI

Pavi Dhiman

Engineer

Research Engineer at Decagon

Decagon team member who wrote about AutoResearch agents that diagnose inference bottlenecks and optimize serving stacks.

Co-author of the Walrus model research paper on cross-domain foundation models for continuum dynamics.

Acknowledged for contributions to Anthropic's work on agent security.

Led earlier Trinity Mini DrugProt-Think project that showed compact open models could learn scientific specialties.

Discussed how AI coding tools dramatically increased code volume at Synthesia, creating massive review bottlenecks.

Extremely productive developer running 5-10 agent sessions concurrently and discusses coding practices with AI agents.

Petr Borovlev is a co-author of the Mellum 2 technical report on the Mixture-of-Experts language model architecture.

Peyton Walters

Engineer

Engineer at Modal Labs

Modal engineer who offered advice and support on sandbox integration for RL training.

Phil Estes, AWS principal engineer, discussed how Cedar was designed as a formally verified authorization system for applications.

Philip Turner

Engineer

Co-created metal-flash-attention implementation that influenced fast matrix multiplication program on Apple GPU.

Author of "Expensively Quadratic: The LLM Agent Cost Curve"

Developer of Heretic, a tool for training-free language model modification with arbitrary-rank ablation, enabling model behavior steering on consumer hardware.

Key contributor to Cohere Transcribe model development.

Pierre Stock

Engineer

AI Scientist at Mistral AI

Mistral AI employee discussed on launch day of Mistral Large 4 regarding training, evals and sandboxing.

Piotr Żelasko

Engineer

Principal Research Scientist at NVIDIA

Co-author on NVIDIA Nemotron 3.5 Transcribe multilingual speech model with transcription and speaker attribution capabilities.

Co-author of Meta Engineering article on thin orchestrator pattern for more reliable AI agents routing to specialists.

Co-author of guide on fine-tuning NVIDIA Cosmos Predict 2.5 with LoRA/DoRA for robot video generation.

Prafulla Dhariwal

Engineer

Technical Fellow at OpenAI

Multimodal Lead for GPT-Image-1.5 project at OpenAI.

Pranay Prakash

Engineer

Vercel

Byline author of article on durable-harness technology for running agent harnesses durably in the cloud.

Author of technical guide on building Duolingo-style AI video call characters using Rive animation platform.

Prashant Kommireddi

Engineer

Engineering Director, AI at Meta

Co-authored Meta's work on AI agents that learn from expert feedback and compile improvements into knowledge bases.

Co-author of OpenAI blog post on the Responses API architecture and its advantages for multimodal and reasoning models.

Authored the NVIDIA MCG Toolkit blog post about automated AI model documentation.

Prava

Engineer

Developer at OpenAI

Co-author of the article and creator of the prava-fc-small grounding model used in Archon

Co-author of CUGA research on agent orchestration and state management.

Made crucial contributions to the MCP Registry project from Last9

Spoke about context engineering at AIEWF, discussing how context flows from business systems to agents.

Member of SGLang team that integrated DFlash and developed Spec V2 engine for speculative decoding.

Co-authored article on Claude Code team's code review process

Qingyang Wu

Engineer

Research Engineer at Together AI

Generated R2E-Gym trajectories for SFT and verifier training, evaluated DeepSWE models, and managed infrastructure.

Quentin Carbonneaux

Engineer

Research Engineer at Meta FAIR CodeGen Team

Contributed to the development and release of Code World Model for advancing code generation with world models.

Rachael Burns

Engineer

Staff Engineer & AI Tech Lead at Oscar Health

Rachael Burns from Oscar Health praised the updated OpenAI Agents SDK for making automation of critical clinical records workflows production-viable.

Principal AI ML Threat Intelligence Engineer discussing AbbVie's use of LLMs for cybersecurity threat detection and analysis

Radek Osmulski

Engineer

Deep Learning engineer at NVIDIA

Co-authored NVIDIA Nemotron 3 Embed embedding models research and release announcement.

Made crucial contributions to the MCP Registry project from Stacklok

Ragavendra Dani

Engineer

Developer

Built GemmaFC, a real-time football simulation controlled by Gemini running on Cerebras infrastructure.

Principal engineer at Microsoft Digital leading multi-agent orchestration for network operations and incident management.

Rahman Syed

Engineer

Sr. Solutions Architect at AWS

Co-speaker at AWS workshop on securing agentic systems and observing AI agent behavior.

Rahul Mohan

Engineer

Senior AI Engineer at You.com

Senior AI engineer reporting significant performance improvement from swapping NVIDIA Nemotron 3 Embed in re-ranking stack.

Rajko Radovanovic

Engineer

Technical contributor at Andreessen Horowitz

Provided targeted technical insights and practical assistance strengthening critical study components.

Co-authored technical post on Agent Compute Engine, Google's foundational infrastructure for running autonomous AI agents at scale.

Rameswar Panda

Engineer

Distinguished Engineer at IBM Research

Key architect of Granite language models, quoted on Granite 4.1's competitive instruction-following and tool-calling performance for enterprise workloads.

Co-authored technical publication announcing Microsoft's Maia 300 chip and rack-scale AI infrastructure.

Raúl Gómez

Engineer

Vercel

Captured the concept that CLI is the fundamental coding agent abstraction

Discussed OpenAI's conservative approach to backend design and how recent model improvements enabled acceleration of chip design tasks

Co-presented measured performance results for Jalapeño accelerator at HotChips 2026.

Developer and contributor to olmOCR project.

Rémi Cadène

Engineer

Former Tesla Scientist at UMA

Unveiled Northstar, a faceless humanoid robot targeting European factories and learning from demonstrations.

Teacher from Texas hired as one of the first super raters for Google's AI products; discusses wage disparities and union organizing efforts

Provided support on inference infrastructure for Chroma Context-1 training.