Write a PRD

Product Requirements Document (PRD)

Feature: Automated Meeting Summaries, Decisions, and Action Item Attribution Product: Huddle Video Conferencing Target Audience: Engineering Leads, Design Leads, Product Ops Status: Ready for Review Target Release: Q3

---

1. Executive Summary & Objective

Today, Huddle’s 3,000 corporate customers rely on manual note-taking. This results in lost context, untracked deliverables, and administrative drag.

This project introduces an AI-powered pipeline that automatically generates: 1. A concise meeting summary. 2. A structured list of key decisions. 3. Assigned action items mapped strictly to verified attendees.

Key Objectives

• Reduce post-meeting admin time by 80%.
• Achieve $\ge 90\%$ user acceptance rate on action item assignments (measured by lack of manual re-assignment or deletion).
• Hard Rule Compliance: 0% assignment of action items to non-attendees.

---

2. Participant Verification & System Constraints

Our models must handle two structural data flaws: 1. Diarization/Labeling Error Rate: ~8% of transcript lines have inaccurate speaker tags (common during crosstalk or multi-person hardware rooms). 2. Ad-Hoc Joiners: ~15% of meetings feature participants who were not on the original calendar invite.

2.1 The "Eligible Assignee" Rule Engine (Non-Negotiable)

``` [Calendar Invite] [Client Telemetry Logs] \ / \ / v v +----------------------------------------+ | Participant Reconciliation Engine (PRE)| +----------------------------------------+ | v [Eligible Assignee Roster (EAR)] - Authenticated Users (User ID + Email) - Verified Guests (Session ID + Display Name) | v +----------------------------------------+ | LLM Extraction & Assignment Pipeline | +----------------------------------------+ | +--> Action assigned to User in EAR? --> ACCEPT | +--> Action assigned to External Party? | v Route to In-Room Proxy (or flag as Unassigned Dependency) ```

Constraint Definition: An action item can only be assigned to an entity present in the Eligible Assignee Roster (EAR) for that specific meeting session.

#### Step 1: Ingest Telemetry at Call Termination (`call.ended` event) Do not rely on the calendar invite as the source of truth for presence. Telemetry establishes physical attendance: * Calendar Attendees ($P{cal}$): Used only as an identity-matching fallback. * Connected Clients ($P{conn}$): Every participant with connection duration $> 60\text{ seconds}$ in the meeting telemetry logs. * Eligible Assignee Roster ($EAR$): Defined strictly as $EAR = P_{conn}$.

#### Step 2: Resolving the 15% Uninvited Attendees * Authenticated Joiners: If an uninvited joiner is logged into a Huddle account, resolve their `userid`, display name, and email directly into the $EAR$. * Guest Joiners (Unauthenticated): If a user joins via guest link without an account: * Capture their client display name and the session ID. * Tag them in the $EAR$ as `Guest: [Display Name]` with `userid = null`. * They are eligible for assignment in the UI, but will require an email address input if notifications are to be sent.

#### Step 3: Mitigating the 8% Diarization Noise Diarization errors mean the speaker label alone cannot be trusted for commitments (e.g., if Bob says "I'll take that," but the audio engine tags Alice). * The LLM prompt must cross-reference explicit verbal attribution ("Jane, can you take that?" $\rightarrow$ "Sure") against semantic context rather than relying exclusively on transcript speaker IDs. * When semantic confidence falls below threshold ($\tau < 0.75$), flag the action item as Unassigned rather than guessing.

#### Step 4: External Dependencies (Out-of-Room Third Parties) If a meeting participant says: "We need legal to review this, I'll have Sarah from Legal look at it," Sarah was not present and cannot be assigned the task. * Enforcement: The system assigns the task to the speaker (the in-room proxy) as: Assignee: Current Speaker (`[Speaker Name]`) Task: "Coordinate with Sarah (Legal) to review [X]." * If the speaker cannot be deduced with high confidence, set `assignee = UNASSIGNED` with a note: `External dependency: Sarah (Legal)`.

---

3. Product & Feature Specifications

3.1 LLM Processing Pipeline

Within 60 seconds of call termination, the processing service executes the following workflow:

``` [Call Ended] | v [Fetch Clean Transcript + Telemetry EAR] | v [LLM Context Construction] ├── System Prompt & Guardrails ├── Strict EAR List (Names + Emails + IDs) └── Diarized Transcript | v [Structured Output JSON Generation] ├── Executive Summary (Paragraph, <= 150 words) ├── Key Decisions (Bullet points) └── Action Items (Array: Task, Assignee ID, Due Date, Evidence Quote) | v [Deterministic Post-Processing Validator] ├── Cross-reference every Assignee ID against EAR └── Overwrite invalid assignments to "UNASSIGNED" | v [Summary Artifact Created & Persisted] ```

3.2 Action Item Post-Processing Validation Logic

Before writing to the database, run a deterministic verification function:

```python def validateactionitems(extractedactions, eligibleassigneeroster): validatedactions = [] for action in extractedactions: assigneeid = action.get("assigneeid") # Validation: Assignee must explicitly exist in the EAR if assigneeid and assigneeid in eligibleassigneeroster: action["status"] = "assigned" action["assignee"] = eligibleassigneeroster[assigneeid] else: # Fallback for hallucinated or external assignments action["status"] = "unassigned" action["assignee"] = None action["flag"] = "ASSIGNEENOTINROOM" validatedactions.append(action) return validated_actions ```

---

4. Sales Request Resolution: Auto-Email Notifications

The Conflict

Sales requested that action items be emailed automatically to assignees immediately when the meeting ends. However, because speaker labels fail ~8% of the time, immediately sending unreviewed emails will blast misattributed commitments to clients and internal executives, eroding trust.

The Solution: "Grace Period Auto-Dispatch"

We satisfy Sales' need for immediate execution while protecting data integrity through a time-delayed trigger and host bypass mechanism.

``` [Meeting Ends] │ ├─► AI Generates Artifacts (Target: <60 seconds) │ ├─► Start 5-Minute Grace Period │ │ │ ├─► Host edits/approves? ──► Send instantly with updates │ │ │ └─► Timer expires? ────────► Auto-dispatch without host intervention │ └─► (Workspace Setting Override: "Instant Send" with 0-min delay available) ```

1. Host Review Window (Default): * A 5-minute countdown starts the moment the summary finishes processing. * The meeting host gets a browser/desktop push notification: "Your Huddle summary is ready. Auto-sending in 5:00 minutes. Click to review." * If the host makes edits or clicks "Send Now", the countdown ends and emails dispatch immediately. * If the host takes no action, the emails dispatch automatically at the 5-minute mark. 2. Sales/Workspace Configuration Flag: * Enterprise Admins can toggle the grace window per team (e.g., Sales team can set Grace Period to `0 minutes` for true instant dispatch; default for other workspaces is `5 minutes`). 3. Safety Fallback: Action items marked as `UNASSIGNED` are omitted from individual assignment emails and instead bundled exclusively into the primary meeting recap sent to the host and organizer.

---

5. User Experience & Design Specs

5.1 Post-Meeting In-App Summary Tab

Located inside the Huddle desktop and web apps under the meeting historical detail view.

``` +-----------------------------------------------------------------------+ | Meeting Summary: Product Architecture Sync [Share] [Edit]| | Today at 10:00 AM • 42 mins • 6 Attendees | +-----------------------------------------------------------------------+ | EXECUTIVE SUMMARY | | The team finalized the data migration strategy for Q3. Key risks | | around API rate limiting were addressed by introducing Redis caching.| +-----------------------------------------------------------------------+ | DECISIONS MADE | | • Chose Redis over Memcached for distributed cache. | | • Deferred mobile client updates to Sprint 44. | +-----------------------------------------------------------------------+ | ACTION ITEMS Auto-sending in [ 04:12 ] [x]| | [ + Add Action Item]| | [x] Task: Implement Redis cache layer | | Assignee: [ (Avatar) David Chen v ] Due: [ Friday v ] | | Source: "David: I can set up the cluster by end of week." | | | | [ ] Task: Verify Legal compliance on data retention | | Assignee: [ Unassigned v ] Due: [ Set Date v ] | | ! Flag: External dependency (Sarah / Legal was not present) | +-----------------------------------------------------------------------+ ```

5.2 Key UI Components

1. Assignee Dropdown Menu: * Filtered exclusively to individuals in the $EAR$. * Section 1: Authenticated Attendees (1-click select). * Section 2: Ad-Hoc / Guest Attendees (Displays `[Guest Name] - prompt for email`). * Option to select "Unassigned". * Explicitly disallows searching the global company directory within the "Assignee" field to prevent violating the non-attendee constraint. 2. Context Tooltip ("Source"): * Hovering over an action item shows the transcript snippet that triggered the assignment, allowing the host to quickly verify ambiguous attributions caused by diarization errors.

---

6. Email Dispatch Specs (Sales Notification System)

6.1 Individual Assignee Notification

• To: Assignee email
• Reply-To: Host email
• Subject: `[Action Required] Action item from: {Meeting Title}`

``` Hi {Assignee First Name},

The following action item was assigned to you during {Meeting Title}:

Task: {Task Description} Due Date: {Due Date or "Not specified"} Context: "{Relevant transcript excerpt}"

View full meeting notes and recording: {Link to Huddle Meeting Record}

Did the AI get this wrong? Click here to reassign or remove: {One-Click Edit Link} ```

---

7. Telemetry, Edge Cases & Error Handling

---

8. Success Metrics

1. Attribution Precision: $\ge 98\%$ of assigned action items correctly reflect real commitments made by that person (measured via sampling and low manual-edit rates).
2. Zero-Hallucination Rate: $100\%$ compliance with the hard constraint: 0 automated emails dispatched to individuals outside the $EAR$.
3. Dispatch Velocity: 95th percentile of summary and action item deliveries completed within $< 90\text{ seconds}$ post-meeting (or exactly at the expiration of the chosen grace period).
4. Host Intervention Rate: $< 12\%$ of summaries require manual host corrections during the grace period window, indicating high trust in baseline accuracy.

Huddle AI meeting summaries

Product requirements for engineering and design leads

Purpose and product decision

Huddle will generate a reviewable record of meeting decisions and actions from its transcripts. The host checks the draft, resolves ownership and publishes it. Approved actions can then be emailed to their owners.

Huddle serves 3,000 companies whose users currently write notes themselves. Speaker labels are wrong about 8% of the time, and roughly 15% of meetings include an uninvited participant. Calendar membership and transcript attribution therefore cannot establish attendance or ownership.

The MVP will not fulfil Sales’ request for immediate, unreviewed owner emails. It will generate drafts automatically after transcription and send emails automatically after explicit publication. Faster delivery does not justify assigning work to absent people. All requirements and numerical targets below are proposed product decisions.

Scope and intended outcome

Primary users are meeting hosts reviewing records and attendees receiving actions. Success means less time producing dependable notes, with no assignments to non-attendees.

MVP includes English transcripts, decisions, action extraction, attendance verification, host review, publication and owner emails. Exclude live summaries, recurring reminders, task-system integrations and automatic reassignment. Use existing meeting access and retention controls.

An action has one verified owner or remains unassigned. Split genuinely separate responsibilities into separate actions. Do not infer deadlines, commitments or decisions from tentative discussion.

Attendance is a release dependency

Engineering must first confirm whether Huddle retains authenticated participant identities and join/leave events. The supplied calendar list is insufficient. If these records are unavailable, build attendance instrumentation before enabling assignments; summaries may still launch with unassigned actions.

An eligible owner must have a stable person identifier linked to an authenticated, human participant session with a recorded join event for this meeting occurrence. Calendar invitees who never joined are ineligible. Bots and room devices are ineligible. Deduplicate reconnects by identity. Someone who left early remains eligible, although a later discussion does not prove they accepted work.

Uninvited users are eligible when their authenticated join is recorded. Guest display names, shared-room speaker labels and email addresses typed by a reviewer are not identity evidence. Unverified guests and people represented only by a room device remain ineligible in MVP; explain how they can join individually with verified identity for future meetings.

This deliberately sacrifices assignment coverage. The invariant is enforceable against verified participation records, not proof of physical human presence behind an account. If literal physical presence is required, that remains an unresolved product constraint and assignments must not launch under a stronger guarantee.

Ownership and extraction rules

The model extracts structured candidates with supporting transcript spans. Each decision contains its outcome and evidence timestamps. Each action contains a task, supporting spans, optional explicit due date and a nullable candidate participant identifier. Preserve the original date wording alongside any normalised date, using the meeting timezone; ambiguous dates stay unset.

Only explicit commitments or clearly agreed assignments qualify as actions. Suggestions, questions and rejected proposals do not. Where later discussion reverses a decision, present the final supported outcome. If the resolution is unclear, flag it for review instead of declaring a decision.

A model candidate is internal metadata, never an assigned owner. Server validation rejects identifiers outside the eligible attendance set before anything is displayed. The review UI starts every owner field as “Unassigned”; reviewers choose an eligible attendee after checking evidence. Do not preselect names from speaker labels, including for “I’ll do it”.

Render ownership only from validated structured fields. Generated task and decision wording must not embed owner claims that bypass these controls. Unsupported person-specific claims are withheld for review. A request that absent Alex should do something can become an unassigned “Confirm responsibility for…” follow-up only when that follow-up was actually agreed; otherwise omit it as an action.

Review and publication experience

The meeting page shows “Preparing summary” after the call, then “Draft needs review”. The host is the default reviewer; existing authorised meeting editors may also review. If no reviewer is available, retain the draft without sending owner emails.

Display separate Decisions and Actions sections. Each item links to its transcript passage and recording timestamp. Offer edit, delete and add controls. Actions show task, owner and due date. An attendance-only owner picker explains why invitees or unresolved guests cannot be selected. Provide a visible “Needs owner” state rather than silently dropping useful work.

Reviewers must confirm each chosen owner and can publish with unresolved actions. The publish dialogue shows recipients and unresolved-action count, with an explicit “Publish and email owners” control. Actions without owners remain visible but produce no email. Keyboard navigation, labelled controls and screen-reader status announcements are required.

Published records show their revision and approval time. Subsequent edits create a new revision. Ownership changes require the same validation and explicit publication, then notify affected owners with a correction. Never silently replace a published record with regenerated content.

Engineering contract and failure behaviour

Use meeting occurrence IDs, not recurring calendar series IDs. Store transcript version, attendance evidence version, model/prompt version, summary revision, item IDs, evidence spans, owner ID, approval actor/time and delivery status. Keep internal candidates separate from approved owner fields.

Process states are waiting for transcript, generating, needs review, published and failed. Trigger generation only after transcript finalisation. Transcript changes invalidate unpublished approvals and create a fresh draft; published revisions remain intact until a replacement is approved.

Enforce owner eligibility server-side on draft writes, publication, exports and notification dispatch. Manual edits and direct API calls cannot bypass it. If attendance evidence is missing, contradictory or unavailable, block ownership and sending rather than falling back to calendar data. Record rejection reasons for support without exposing transcript content in routine logs.

Use transactional publication and an outbox. Delivery keys combine meeting, published revision and recipient so retries do not duplicate emails. Email only approved actions owned by that recipient, with a permission-checked meeting link. Resolve addresses from verified identities, never model output. Block delivery if permissions or identity validation fail; show the reviewer the failure. Retry transient failures with backoff for up to 24 hours, then offer manual retry.

Treat transcripts as untrusted data: spoken instructions cannot change permissions, choose recipients or trigger tools. Generation has no email capability. Preserve tenant isolation and apply meeting deletion and retention policies to summaries and queued notifications.

Acceptance scenarios

1. Invited Pat never joins but is named in an action: Pat is absent from the picker; API assignment is rejected; no owner email is sent.

1. Uninvited authenticated Lee joins: Lee is eligible and can receive an action after reviewer confirmation and publication.

1. Several people join through one room device: no individual becomes eligible from its label; actions remain unassigned unless independent authenticated participation exists.

1. Overlapping speech mislabels a commitment: no owner is preselected; the reviewer can inspect evidence and choose a verified attendee or leave it unassigned.

1. No decisions or actions are supported: show explicit empty states rather than inventing content. Failed or incomplete transcription shows a retry state and sends nothing.

1. A forged owner ID, missing attendance record or duplicate publish event cannot create an invalid assignment or duplicate email. Revoked access blocks queued delivery.

Evaluation and release

Build a consented, de-identified evaluation set of at least 200 meetings, deliberately including overlap, shared rooms, absent invitees, uninvited guests, duplicate names, reversals and ambiguous commitments. Human annotators label attendance, final decisions, actions, ownership and evidence; adjudicate disagreements independently. Report each difficult slice separately.

Release gates: zero non-attendee assignments or deliveries across end-to-end and adversarial tests; at least 95% precision and 85% recall for supported decisions and actions, measured separately. Zero observed violations is necessary but insufficient: eligibility enforcement must also pass API and notification-path tests. Model confidence alone never permits assignment.

Pilot with ten opt-in companies before staged rollout. Target draft availability within two minutes of final transcript for 95% of meetings up to 60 minutes, and a 50% reduction in median note-preparation time against pilot baseline. Track owner corrections, unassigned-action rate, publication rate and delivery failures. Any non-attendee assignment disables assignments and sending pending investigation. Expand only after two weeks without a critical violation and acceptable review effort; do not reduce abstention merely to improve coverage.

PRD: AI Meeting Summaries — Decisions & Action Items

Sep 24, 2026 · @Martin Slaney

Summary

We will generate a summary after every recorded Huddle meeting that lists the decisions made and the action items, each with an owner who was verifiably in the meeting. The organizer reviews the summary and sends it in one click; nothing is emailed without that step in v1.

Problem. Today attendees write their own notes after the call. Notes are inconsistent, often missing, and action items lose their owners within a day. We already have what we need to fix this: a recording, a transcript with speaker labels, and the calendar attendee list.

Why this is not trivial. Two of those inputs are unreliable. Speaker labels are wrong about 8% of the time, mostly during crosstalk and when several people share a meeting-room device. About 15% of meetings include someone who joined without a calendar invite. A naive build would misassign owners often enough to lose trust in the first week.

Goals and metrics

The goal is summaries people send without rewriting, with zero actions assigned to non-attendees.

Goals

• Replace manual post-meeting notes for recorded meetings.
• Every action item has a correct owner or is explicitly marked unassigned.
• Get action items to owners within minutes of the meeting ending.

Non-goals (v1)

• Pushing tasks into Jira, Asana or Linear.
• Summaries for unrecorded meetings or live, in-call summaries.
• Improving diarization itself. We design around the 8% error rate, not fix it.

Success metrics

Owner precision is measured from organizer edits plus a weekly labelled sample of 200 meetings.

Users and core flow

The organizer owns the summary; attendees receive it.

• Organizer: reviews, fixes owners, sends. Usually the calendar invite's creator; can hand off to any attendee.
• Action owner: receives their items by email and in the app.
• Other attendees: receive the full summary once sent.

```mermaid flowchart LR A[Meeting ends] --> B[Transcript +<br/>join log ready] B --> C[Build verified<br/>roster] C --> D[Extract decisions<br/>and actions] D --> E[Roster filter +<br/>confidence check] E --> F[Organizer review] F --> G[Send to owners<br/>and attendees] ```

The roster filter runs as deterministic code after the model, never inside the prompt alone.<sup>2</sup> Target: summary ready for review within 5 minutes of meeting end for meetings up to 90 minutes.

Requirements

An owner can only be chosen from the verified roster: people who actually joined, not people who were invited.

Verified roster

1. R1. Build the roster from the join log, not the calendar invite.<sup>1</sup> Invited people who never joined are excluded. 2. R2. Each roster entry has an identity tier: - Verified: joined signed in to a Huddle account, or by a dial-in matched to a known number. - Guest: joined with a display name only (most of the 15% uninvited joiners). - Room: a meeting-room device; people behind it are unknown individuals. 3. R3. Match uninvited joiners to workspace accounts by signed-in account only. Never match on display name alone.

Summary content

1. R4. Decisions: a short statement of each decision, with a transcript timestamp link. No owner field.
2. R5. Action items: task, owner (or "Unassigned"), due date if said aloud, timestamp link, and confidence.
3. R6. Every item links to its evidence in the transcript so reviewers can check it in one click.

Assignment rules

1. R7. The model returns a proposed owner per item. Code then checks it against the roster. Any owner not on the roster is replaced with "Unassigned", and the named person is shown as a mention ("mentions: Dave, not in meeting").
2. R8. Owner comes from content first: a named request ("Priya, can you…") beats the speaker label of "I'll do it".
3. R9. Mark an owner low confidence when the evidence depends on a speaker label in an overlapping segment, a Room entry, or a Guest entry.
4. R10. Low-confidence items show as "Suggested: Priya" and need organizer confirmation before any email goes out.
5. R11. Room and Guest entries cannot receive email. The organizer can reassign to a verified person or leave it unassigned.
6. R12. Log every model proposal, filter result and organizer edit. This is our precision measurement.

Handling the known limits

Both limits hit ownership, not decisions, so we contain them at the owner step.

The error compounds per meeting. If each item's owner is wrong 8% of the time, a meeting with 5 action items has a 34% chance of at least one wrong owner (1 − 0.92⁵). That is why confirmation sits before email, not after.

Sales request: auto-email on meeting end

Decision: not in v1 as requested. We ship one-click organizer send now, and gated auto-send in phase 2 once measured precision earns it.

Why not now. An email to the wrong owner is the most visible failure this feature can have. With the error rate above, roughly a third of 5-item meetings would email someone a task that isn't theirs. With external guests, it may land outside the customer's company. Early trust is hard to win back.

What Sales gets instead.

• v1: the organizer gets a "Summary ready" notification within 5 minutes and sends with one click. Goal: sent within 30 minutes (see metrics).
• Phase 2: a workspace setting to auto-send at meeting end, but only items that are high confidence and owned by a Verified attendee. Everything else still waits for the organizer. It ships only when owner precision on high-confidence items is ≥ 97% for 4 consecutive weeks.
• Auto-send never emails Guest or Room entries, or anyone outside the customer's domain, unless an admin enables it.

Design notes

The review screen must make checking owners faster than writing notes.

• Decisions first, then action items grouped by owner, then Unassigned.
• Low-confidence owners look visibly different ("Suggested") and are editable inline from a roster picker, never free text.
• Each item plays its transcript clip on click.
• The Send button states who gets email: "Send to 4 owners and 7 attendees".
• Empty state: if no decisions or actions are found, say so plainly rather than inventing any.

Rollout

Open questions

• [ ] Is the join log's signed-in identity reliable enough to call "Verified"? Engineering to confirm.
• [ ] Who organizes when the invite creator didn't attend? Proposed: first Verified attendee from the invite.
• [ ] Do customers need to disable summaries per meeting for sensitive calls? Legal and Sales to confirm.
• [ ] Retention: how long do we keep drafts and the evidence log?