About Beyalan
An AI-native investment research platform — describe what you want, and the analyst builds the screen, not just the answer.
Our Mission
Beyalan is AI-native research, not AI-assisted software. The analyst agent is the primary user of every capability — markets, fundamentals, macro, and quant — and the interface is an increasingly generated surface: you ask in plain English, and Beyalan gathers live data, composes an interactive GUI, and chains it into repeatable workflows.
The same quantitative tooling once locked inside hedge funds — ~150 research tools, a 10-year daily valuation database, factor and regime analytics, financial models — is here, free to use and with no quant background required. And because the durable IP is the learning loop layered on top of the models, every use compounds into owned, model-agnostic expertise rather than evaporating.
How it works
Every view flows through one three-stage pipeline.
Gather
Deterministic where the shape is known (fast, exact), agentic where it's open-ended — always live data, never invented.
Compose
An AI pass turns the gathered data into typed UI blocks — metrics, tables, charts, the read-through.
Render & own
Deterministic React renders it. Pin views, chain workflows, export to branded Excel & PDF — the loop is yours.
See the platform
The agentic surfaces (Studio, analyst, stock agent) alongside the deterministic dashboards and integrated AI insights.
CPI cooled to 3.1% YoY; shelter still sticky. The disinflation path supports the soft-landing read — watch core services.
Connected to everything
Brokerage, market, fundamentals, and macro data — plus the AI and platform layer — all wired into one surface.
Market & fundamentals
Macro
Brokerage
AI
Platform
The same tools power the in-app analyst and any external agent over the MCP server — read-only, RLS-scoped, bring-your-own-key.
Leadership
Dzhan Yusnyu
Founder & CEO
Dzhan Yusnyu is an investment professional with over five years of experience managing discretionary macro and systematic tactical asset allocation strategies across a universe of U.S. listed global equity and fixed income ETFs. He holds additional expertise in creating and managing single stock strategies — both systematic (quantitative, factor-based, and fundamental) and discretionary (thematic).
Prior to founding Beyalan, Mr. Yusnyu served as Senior Analyst at Horizon Investments, where he supported portfolio implementation decisions across a broad range of strategies, conducted macroeconomic research, and evaluated products and methods to express investment views across strategies. Before Horizon, he held analyst roles at Blackford Capital, a Michigan-based private equity firm.
Mr. Yusnyu is well versed in portfolio analytics, risk attribution, manager diligence, risk management, and all other aspects of institutional portfolio management. He leverages generative AI to streamline research and portfolio management workflows — converting Excel models, Python scripts, and other error-prone processes into highly scalable and universally interactive applications.
Mr. Yusnyu graduated from the University of North Carolina at Chapel Hill with a B.A. in Economics, a B.A. in History, and a minor in Politics, Philosophy, and Economics. He also holds a Credential in Quantitative Financial Economics from UNC-Chapel Hill and is a CFA Level II candidate.
The Platform
Studio — generative GUI
Describe a view in plain English and the agent composes a live, interactive screen. New views are recipes, not hand-built pages.
Workflows & dashboards
Chain steps into an end-to-end research process the agent drafts and you edit; compose boards of live tiles that re-run on open.
Analyst & stock agents
A tool-using analyst (~150 tools) that reasons across markets, fundamentals, macro and quant — and deep-dives any single name.
Portfolios & brokerage sync
Connect a brokerage read-only via SnapTrade, then track TWR/MWR, factor exposures, and concentration risk with AI-read analytics.
A learning loop
Every run is captured as a trace; your up/down ratings promote the best into RAG-injected institutional memory. Private evals, a model-swap sovereignty test, and agent-proposed, human-approved upgrades keep quality compounding — the loop, not the model, is the IP.
Team workspaces
Invite your desk by email to share institutional memory and approved upgrades across the team. What one analyst validates, everyone inherits — per-user work stays private by default.
Agent-ready over MCP
The same tools — and the same gather→compose loop — that power the in-app analyst are exposed to any external agent over the MCP server, read-only and RLS-scoped.
Get in Touch
Questions, feedback, or partnership inquiries? Reach us at help@beyalan.com