| ICLR 2024, Part 2 of 5: Compute, Context, and Benchmarks for Working Agents …2024, Part 2 of 5: Compute, Context, and Benchmarks for Working Agents Part 2 moves from the experiments academia can afford to run into SWE-Bench, contamination detection, and GAIA’s tests of everyday assistance. Writeup · Date unavailable | Writeup | — | — |
| ICLR 2024, Part 4 of 5: Verifiers, Safety Systems, and Agents ICLR 2024, Part 4 of 5: Verifiers, Safety Systems, and Agents Part 4 follows stepwise reasoning feedback into deployment safety, browser automation, and multi-agent software development, then closes with the reversal curse: why learning a… Writeup · Date unavailable | Writeup | — | — |
| Language Agents: From Reasoning to Reliable Action Language Agents: From Reasoning to Reliable Action Reasoning helps an agent choose its next move. Reliable action also depends on usable tools, meaningful feedback, memory, and control over the workflow. Writeup · Date unavailable | Writeup | — | — |
| Lindy at Work: Putting Agents on Rails Lindy at Work: Putting Agents on Rails Lindy’s redesign makes required actions explicit while leaving models room to interpret information—a division of labor that shapes its workflows, memory, evaluations, and product strategy. Writeup · Date unavailable | Writeup | — | — |
| Notion’s Custom Agents: Rebuilding the System Around the Model Notion’s Custom Agents: Rebuilding the System Around the Model Background work requires understandable permissions, dependable tools, useful evaluations, and context that extends beyond a chat. Writeup · Date unavailable | Writeup | — | — |
| One Year of MCP: From Local Tools to Asynchronous Agents One Year of MCP: From Local Tools to Asynchronous Agents MCP’s first year exposed the hard parts of shared agent infrastructure: enterprise identity, bidirectional transport, trusted discovery, long-running work, and neutral governance. Writeup · Date unavailable | Writeup | — | — |
| Serverless Browsers and the Infrastructure Behind Computer-Using Agents Serverless Browsers and the Infrastructure Behind Computer-Using Agents Browserbase combines hosted browsers, natural-language automation and human takeover to handle the state, access and reliability problems beneath an agent’s clicks. Writeup · Date unavailable | Writeup | — | — |
| Shipping agents means building the application around them Shipping agents means building the application around them Durable workflows, streaming interfaces, operational context, and external security controls turn an agent loop into a usable application. Writeup · Date unavailable | Writeup | — | — |
| The RLVR Revolution: From Checking Answers to Training Agents The RLVR Revolution: From Checking Answers to Training Agents Tulu’s practical post-training recipe leads to harder questions: how models learn to use tools, choose plans, recognize success, and remain useful as rewards become more complex. Writeup · Date unavailable | Writeup | — | — |
| What It Took to Replace a Conference SaaS Workflow With AI Agents What It Took to Replace a Conference SaaS Workflow With AI Agents A speaker-management contest tested AI-built software against the details of real event operations—and exposed the gap between a demo, an evaluation and an owned service. Writeup · Date unavailable | Writeup | — | — |
| Why agents need reasoning, inspectability, and better abstractions Why agents need reasoning, inspectability, and better abstractions Imbue’s approach starts with a recruiting product’s trust problem and asks what it takes to make models dependable enough to delegate work to. Writeup · Date unavailable | Writeup | — | — |
| Long Context, Working Agents, and Learning in Public — Part 2 of 3 Long Context, Working Agents, and Learning in Public — Part 2 of 3 Part 2 of the January–February 2024 recap: Gemini’s million-token context, prompt transparency, anniversary demos of Lindy, RWKV, Pixee, and Julius, and listeners on… Writeup · Date unavailable | Writeup | — | — |
