Opportunities

Coding agents now produce a growing share of a team's code, while the reasoning behind each change -- the alternatives weighed, the constraints discovered, the

Safe physical AI for robot actions are required not only likely to succeed but tested to be safe before execution. In practice, however, formal testing of motio

As Large Language Models (LLMs) evolve into autonomous agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipe

Frontier LLM agents are automating many business tasks, but their high inference cost makes large-scale deployment unsustainable. Small language models (SLMs) o

Large language models (LLMs) are increasingly used in agentic coding settings, where they can inspect files, execute commands, run tests, observe failures, and

LLM agents often retry after external validation rejects a candidate, but the interface between validation and the next model call remains underspecified. We in

While autonomous coding agents have significantly advanced automated test generation, they remain fundamentally limited by lazy generation, a phenomenon where a

Show HN: I built a web tool to see and edit what an AI thinks before it answers

Show HN: StyleSeed – a design-rules engine so AI agents stop building generic UI

An Enterprise Permission Layer for AI Agents

Zero-dependency browser video editor that AI agents can drive — JSON timeline, MCP + REST, live-reloading UI

From-scratch Rust+CUDA inference engine, bit-exact by construction — NVFP4, MoE, MTP speculative decoding, tuned against measured limits of one RTX 5090 Laptop

RL-training an AI agent to RL-train AI agents.

LLM agents as your hyperparameter optimizer.

Live voice + vision AI assistant — on-device voice pipeline, thin LLM proxy

Creating an AI agent with LISP - the 2026 version of what made LISP "the language for AI" in the first place

Repo-native shared memory for AI coding agents.

The broad adoption of Artificial Intelligence (AI), especially Generative AI, raises pressing questions about how users interact with these systems to produce n

Слойка — AI-first графический редактор (Photoshop-style, WebGL2, SAM 2.1, polza.ai, Claude Code/MCP)

Open-source natural-language analytics platform that converts business questions into safe, explainable SQL.