AI Engineer

When Will The Benchmaxxing Plague End? — Nick Heiner, Surge AI

Teaching AI to Find Real Vulnerabilities — Prof. David Brumley, Bugcrowd

Teaching AI models to hack requires a structured approach mirroring human learning, where tasks increase in difficulty from toy problems to hardened targets. Effective reinforcement learning environments must avoid simplistic crash-based grading, which encourages reward hacking and stunts model growth. Implementing an ...

Rethinking Environments for Long-Horizon Work — Rayan Garg, Theta Software

Long-horizon AI agents require a nuanced definition of time horizons that moves beyond simple token counts or human-centric benchmarks. Effective evaluation of these agents necessitates assessing environment complexity, specifically regarding sequential tool coordination and state changes, rather than just parallelizab...

What's Next After RLHF? — Diogo Almeida, TypeSafe AI

RLHF (Reinforcement Learning from Human Feedback) currently dominates the AI landscape, yet it fundamentally limits the potential for true automation. By optimizing models to prioritize human preferences and engagement, developers have created systems that excel at assistance but struggle with autonomous, high-stakes d...

Data Quality Is the Compute Multiplier — Ari Morcos, DatologyAI

Learning on the Job: The Future of Post-Training — Raymond Feng, Applied Compute

Post-training methodologies for AI agents are evolving from simple, controlled Q&A tasks toward autonomous skill acquisition in complex, real-world environments. Current frameworks utilize reinforcement learning, specifically GRPO, to optimize model performance within synthetic sandboxes, though this approach faces sig...

Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs

Scaling to Long Horizons — Ross Taylor & Chengxi Taylor, General Reasoning

Emulated: The Data for Fully Autonomous Software Engineers and Companies — Joseph Wang

The Base Model Is Dead — Varun Singh, Arcee AI

The traditional concept of the "base model," built primarily on massive web-text scrapes to reflect human knowledge, is evolving into a foundation for reasoning and agentic behavior. Modern training paradigms now prioritize reinforcement learning (RL) as a core component rather than a supplementary refinement, with com...

Verifiable Environments for AI in Biology — Kenny Workman, LatchBio

Ending AI Slop — Thais Castello Branco, Taste Labs

Subjective domains like design and creative writing pose significant challenges for AI because they lack the clear, verifiable ground truth found in coding or mathematics. Solving these problems requires decomposing fuzzy concepts into codifiable, measurable components, such as specific brand guidelines for visual asse...

Benchmarks: The Good, the Bad, and the Ugly — Ali Khial, G2i

Reinforcement Learning without Verifiable Rewards — Will Brown, Prime Intellect

Agents at Scale: Inside MiniMax's Model and the Infrastructure Behind It — Dan Fu and Olive Song

fighting slop with slop — Vaibhav Gupta, Boundary

First Steps Toward Automated AI Research — Richard Socher, CEO Recursive AI

Your Finance Agent's Bottleneck Is You — Ramana Siddanth Emani, Auditoria AI

Build for the Memo, Not the Demo — Shawn Chan, China Resources Holdings

Let's integrate AI Agents in Event-Sourced Systems — Divakar Kumar, FlyersSoft