Insights on AI agents, automation, and the future of work from the team building Decisional.
AI intelligence is jagged—superhuman at coding and research, stumbling at basic tasks. This post proposes five principles for building products that ride those edges without pretending the gaps don't exist.
"Just adding AI" has burned a lot of founders, PMs, and enterprises. This post proposes five principles that define what it means to be "AI Native" - from using AI along jagged edges to achieving 100% workflow completion.
Agentic automation with tools outperforms computer use by 10x. METR's research shows AI agent workflows struggle with long action sequences—making agentic process automation with tool-first design the path to reliable agents.
An AI coding assistant accidentally deleted all files on our Mac due to a path confusion error. Without strong guardrails, convenience can outweigh safety in a single command.
A new way for Financial Professionals to do in-depth work. Decisional has launched Knowledge Engines, an AI-native workspace designed for financial professionals working in private markets.
The future of document workflow automation isn't about better retrieval—it's about agents that think, reason, and complete work. RAG isn't dead; it's just beginning to show what's possible.
OpenAI's launch of AgentKit & ChatKit and what it means for AI automation. Analysis of the AI startup landscape and the future of business applications.
Exploring how AI-native workflows are reshaping professional work, drawing insights from OpenAI's approach to building AI-native engineering teams and processes.
How agentic automation flexes to handle complex, variable workflows that traditional rule-based automation cannot. Building systems that adapt at runtime.
Evaluating the top AI-native workflow automation tools of 2025. Unlike traditional rule-based automation, AI-native tools incorporate runtime reasoning and agent-based automation.
Picking the right AI tool for Financial Analysis can save each financial professional up to 30 hours per week. This comprehensive guide helps financial services firms understand AI tools for embedding into their workflows.
Lessons from building an AI agent for financial research. Modern RAG systems have evolved beyond simple retrieval to mirror how human analysts conduct deep research.
The art of allocating intelligence. How AI agents are reshaping the role of knowledge workers and enabling new forms of human-AI collaboration in professional services.
If you run an LLM in a loop and most LLMs are similar, what makes different agents have different capabilities? The answer lies in tools, workspaces, and specialization.
Richard Sutton thinks LLMs are a dead end for AGI. His arguments are valid but hackers and engineers have pragmatic solutions. Exploring the future of AI development.
A battle test of all the deep research tools available out there. We put the leading AI research tools head-to-head on real financial analysis tasks.
Could it be that the AI assisted Financial Professional is already moving markets by $Trillions? Exploring the hidden impact of AI-augmented decision making in finance.
There have been 21 mentions of Deepseek in earnings calls so far but if you really want to know what is going on - listen to builders using AI. Why model agnosticism is essential from day one.
Decisional achieved a 4.1x improvement in financial AI performance—jumping from 29/150 to 119/150 on Financebench—through strategic optimizations in retrieval-augmented generation (RAG) and tool design.
Introducing Decisional, a Y Combinator and NVIDIA Inception-backed startup building AI agents for financial services. While technology has disrupted financial services externally, internal business processes remain largely unchanged.
Subscribe to Neural Markets for insights on AI agents, automation, and the future of financial technology.
Subscribe on Substack