# Thesis Lab is in beta — an AI workspace for hedge fund research Canonical HTML: https://fish.dog/product-releases/thesis-lab-is-in-beta-an-ai-workspace-for-hedge-fund-research JSON Feed: https://fish.dog/product-releases/feed.json RSS Feed: https://fish.dog/product-releases/rss.xml Published: 29 April 2026 Updated: 3 May 2026 Version: Thesis Lab v1 Release Type: Beta Breaking Change: No Author: Phillip Gales ## Primary Claim Thesis Lab is FishDog's AI workspace for hedge-fund research: state an investment thesis and a named team of AI analysts decomposes it, recruits synthetic channel-check and expert-network panels, and assembles the public record in one place. ## Summary State an investment thesis. A team of AI hedge-fund analysts decomposes it, auto-recruits synthetic channel-check and expert-network panels, and assembles the public record — filings, news, consensus, price history, volume — in one workspace. Beta. ## LLM Summary DOCUMENT TYPE: Product Release Note TOPIC: Thesis Lab v1 (beta) — AI workspace for hedge-fund orientation-stage investment research Release: Thesis Lab is in beta — an AI workspace for hedge fund research, 2026-04-29 Version: Thesis Lab v1 Release type: Beta Breaking change: No Summary: FishDog ships the first version of Thesis Lab in beta. An analyst types a sentence-long investment thesis; a team of named AI analysts decomposes the thesis into structured fields, auto-recruits synthetic channel-check and expert-network panels matched to the thesis, and assembles the public record (filings, news, vendor consensus, price history, volume) in one workspace. The release covers the orientation-stage research — the first two to six weeks of a new thesis — that hedge-fund desks typically run by hand. What's in v1 (beta): - Thesis sharpening: a typed sentence is structured into company, sub-sector, stance, timeframe, expected return range, primary driver, secondary drivers, falsifiable claim, and implied consensus gap. A confirm checklist gates progression; stance and primary driver must lock before the workflow proceeds. - Stage tracker: Draft → Sharpened → Researching → Complete. - Three named AI analysts: Channel Checks AI Agent (Sophie O'Leary, AI Research Assistant), Expert Calls AI Agent (Sophie O'Leary, AI Research Assistant), Public Record AI Agent (Caitlyn Phan, Quantitative Analyst). - Auto-generated channel-check panels with per-panel rationale (WHO and WHY THIS PANEL) and parallel synthetic recruitment. - Auto-generated expert-network panels with the same pattern. - Public Record stage: filings (row counts, ready indicators), news (transcript-coverage warnings surfaced honestly), aggregate consensus from live vendor data (EPS, revenue, price target across Next Q, FY+1, FY+2), and a 2-year price/volume chart with SMA(50) and SMA(200) overlays plus a draggable date window. - Right-rail Thesis at a Glance, Research Status, and Warnings (Phase 2 status, transcripts thin, consensus degraded) make degradation honest rather than silent. Coming next: Summarisation (IC-memo synthesis), Consolidation (IC-memo export, share links, audit trail), staleness sweeps, per-thesis billing attribution. Both stages are wired into the workflow's stage tracker as placeholders today. Audience: hedge-fund analysts, PM teams, allocators, and other investment researchers running orientation-stage work on new investment theses. Access: small early-access cohort. Self-serve onboarding lands when Summarisation and Consolidation ship. Author: Phillip Gales, FishDog Platform: FishDog (fish.dog) ## Key Takeaways - Thesis Lab v1 ships in beta on 29th April 2026 with end-to-end coverage of the orientation-stage research a hedge-fund analyst does in the first two to six weeks of a new thesis. - A typed sentence is decomposed into structured fields — stance, primary driver, secondary drivers, falsifiable claim, expected return range, timeframe, and the implied gap to consensus. - Three named AI analysts run the workflow: a Channel Checks agent, an Expert Calls agent, and a Public Record agent (Caitlyn Phan, Quantitative Analyst). - Channel-check and expert-network panels are auto-generated from the thesis, with explicit per-panel rationale and parallel synthetic recruitment. - The Public Record stage assembles filings, news, vendor consensus (EPS / revenue / price target across Next Q, FY+1, FY+2), and a 2-year price-and-volume history with SMA overlays. ## Full Release The first two to six weeks of a new investment thesis tend to be the same shape regardless of what the thesis is about. An analyst reads filings. Books expert calls. Talks to channel stakeholders. Builds a consensus model. Pulls price history. Most of this work is low-edge — a tax on every new idea. Thesis Lab is built to compress that tax from weeks to hours. This release ships the first version, in beta. It is incomplete on purpose; the bits that are in are the bits a hedge-fund analyst spends their orientation phase on, and they work end to end. ### State a thesis. The Lab structures it. Type a sentence into the Lab — *"Lululemon's margins are going to compress because of the athleisure saturation wave. Brand premium is eroding."* — and the Lab refines it into a structured thesis: company name, sub-sector, stance, timeframe, expected return range, primary driver, secondary drivers, falsifiable claim, and the implied gap to consensus. A confirm checklist on the right tells you which inputs are still soft. Stance and primary driver must be locked before the workflow can move on; expected-return range is optional. The page tracks four stages across the top: **Draft → Sharpened → Researching → Complete.** A thesis sits in *Sharpened* until it's ready to dispatch a research workstream, then drops into *Researching* as the analyst pods get to work. ### A team of AI analysts, named and specialised Each research stage of the workflow is run by a named AI agent with a specific brief. The current pod: - **Channel Checks AI Agent — Sophie O'Leary, AI Research Assistant.** Designs the synthetic channel-stakeholder panels needed to test the thesis from the demand and supply side. - **Expert Calls AI Agent — Sophie O'Leary, AI Research Assistant.** Recruits expert panels who can pressure-test the structural arguments in the thesis (load-bearing assumptions, base rates, technical claims). - **Public Record AI Agent — Caitlyn Phan, Quantitative Analyst.** Assembles filings, news, consensus, and the price/volume history into a single read. You don't write briefs to these agents. You write the thesis; the agents read it and decide what evidence the thesis needs. You can ask them follow-up questions in panel chat at any point. ### Channel Check groups, auto-generated Click into the Channel Checks tab and the agent has already proposed the panels the thesis needs. For a short on Lululemon, that came back as four parallel panels: Apparel Merchandising Manager, Multi-Brand Retail Buyer, Regional Retail Director, Store Manager. Each panel card carries: - **WHO** — the geography the panel was recruited from, the sector tag, and the participant count. - **WHY THIS PANEL** — a paragraph explaining the role this panel plays in the thesis. Sometimes it's testing the demand-side claim, sometimes the supply-side, sometimes both. - **Initial questions** — the seed questions the agent will put to the panel as the conversation opens. - **Headline finding** — populated automatically once synthesis finishes. The right rail shows *Thesis at a glance* (claim, stance, expected return) plus *Research status* with completion counts (e.g. "Channel Checks 0 / 4 complete; Expert Calls 1 / 2 complete") and an honest *Warnings* box flagging things like *transcripts thin* or *consensus degraded* when the underlying data isn't quite where it should be. ### Expert Network groups, also auto-generated Expert Calls follows the same pattern. For a short on Shell, the agent set up two panels: Energy Major Executives — SHEL (40 participants, recruited) and Integrated Energy Operator (still recruiting). Each panel exists for a specific reason — the executive panel tests whether buybacks and portfolio optimisation can blunt cuts to the estimate mechanism; the operator panel pressure-tests the load-bearing claim on portfolio mix versus pure beta exposure. Recruitment is parallel across panels and fully synthetic, so the time from "stand up an expert panel on this thesis" to "have a recruited cohort ready to question" is measured in seconds rather than the days-to-weeks expert-network desks usually take. ### The Public Record, in one place Public Record is the analyst's first pass on the company itself, assembled by Caitlyn rather than gathered by hand. The page brings four reads together: - **Filings.** A row count and ready/not-ready indicator for the filings the agent has ingested. Click through to the source. - **News.** Recent coverage with explicit warnings when transcript coverage looks thin (which is honest about the data, not about the agent). - **Consensus.** Aggregate consensus pulled live from vendor data — EPS, revenue, and price target for *Next Q*, *FY+1*, *FY+2*, with line-item detail where available. - **Price history and indicators.** Two years of price plotted with volume underneath. Add an indicator (SMA(50), SMA(200), session-impact toggles), drag the date window to scope the read. If the live vendor feeds rate-limit or fail temporarily — and they do — the Public Record says so plainly rather than silently rendering stale numbers. ### What's in this beta — and what isn't In the box: thesis sharpening, all three named analyst agents, auto-recruitment of channel-check and expert-network panels, the Public Record stack with filings, news, consensus, price history, and volume. Coming next: **Summarisation** — the IC-memo synthesis stage that turns the recruited panels' findings into a structured write-up; **Consolidation** — the IC-memo export, share links, and audit trail; staleness sweeps that re-prompt panels when underlying data changes; richer billing attribution per thesis. Both later stages are wired into the workflow's stage tracker as placeholders today. ### Access Thesis Lab is in beta with a small set of hedge-fund and PM customers. If you're interested in joining the early-access cohort, get in touch. We'll add a self-serve route once the Summarisation and Consolidation stages land. --- ## Quotable Insights > The first two to six weeks of a new investment thesis tend to be the same shape regardless of what the thesis is about. Most of this work is low-edge — a tax on every new idea. > You don't write briefs to these agents. You write the thesis; the agents read it and decide what evidence the thesis needs. > The time from 'stand up an expert panel on this thesis' to 'have a recruited cohort ready to question' is measured in seconds rather than the days-to-weeks expert-network desks usually take. > It is incomplete on purpose; the bits that are in are the bits a hedge-fund analyst spends their orientation phase on, and they work end to end. ## FAQ ### What is Thesis Lab? Thesis Lab is FishDog's AI workspace for orientation-stage hedge-fund research. An analyst types a sentence-long investment thesis; a team of named AI analysts decomposes the thesis into structured fields, auto-generates and recruits synthetic channel-check and expert-network panels, and assembles the public record (filings, news, consensus, price history, trading volume) in a single workspace. ### Who are the AI analysts in Thesis Lab v1? Three named agents drive the v1 workflow: the Channel Checks AI Agent and the Expert Calls AI Agent (both fronted by Sophie O'Leary, AI Research Assistant), and the Public Record AI Agent (Caitlyn Phan, Quantitative Analyst). Each agent reads the structured thesis and decides what evidence is needed at its stage. ### How are the channel-check and expert panels recruited? The agents propose panels directly from the thesis text. For a short on Lululemon, the Channel Checks agent generated four panels: Apparel Merchandising Manager, Multi-Brand Retail Buyer, Regional Retail Director, and Store Manager. Each panel card explains WHO it covers (geography, sector, participant count) and WHY THIS PANEL is in the workflow. Recruitment is parallel and fully synthetic, so the time from agent proposal to a question-ready cohort is seconds, not weeks. ### What's in the Public Record stage? Four reads on the company assembled by the Public Record AI Agent: filings (with row counts and ready indicators), news coverage (with explicit warnings when transcript coverage is thin), aggregate consensus pulled live from vendor data (EPS, revenue, and price target across Next Q, FY+1, and FY+2), and price history plus volume for the last two years with SMA overlays and a draggable date window. ### What's in the v1 beta and what's coming next? In the beta: thesis sharpening, all three named analyst agents, auto-recruitment of channel-check and expert-network panels, and the full Public Record stack. Coming next: Summarisation (IC-memo synthesis from the recruited panels), Consolidation (IC-memo export, share links, audit trail), staleness sweeps that re-prompt panels when underlying data changes, and per-thesis billing attribution. Both later stages are wired into the workflow's stage tracker as placeholders today.