Write a PRD

PRD: AI-Generated Meeting Summaries with Decisions and Action Items

Audience: Engineering and Design leads Status: Ready for build planning Owner: Product

1. Summary

Huddle records and transcribes meetings with speaker labels, and the attendee list comes from the calendar invite. Today, people write their own notes after calls, and those notes are often late, incomplete, or missing. This feature generates a summary after each meeting that lists the decisions made and assigns action items to attendees. Action items are emailed to their owners, with Sales' requirement for immediate delivery balanced against the accuracy limits described in Section 4.

2. Goals and Non-Goals

Goals - Produce a summary within 5 minutes of meeting end for 95% of meetings. - Every action item names an owner who was present in the meeting. - Every decision and action item links to the transcript moment that supports it. - Give Sales automatic, immediate delivery for high-confidence action items.

Non-Goals (v1) - Calendar write-back or task creation in external tools (Jira, Asana). Design for it, but do not build it. - Summaries for meetings without a transcript. - Sentiment, coaching, or participation analytics.

3. Definitions

• Attendee: A person present in the meeting, determined by the participant session log (the platform's record of who joined), not only by the calendar invite. The calendar invite is a hint, not proof of attendance.
• Identified participant: A participant tied to a Huddle account or a verified email address.
• Unidentified participant: A guest who joined by name only, with no verified identity.
• Action item: A task with an owner, a description, and optionally a due date, derived from transcript content.
• Decision: A statement that the group agreed on a course of action, with supporting transcript evidence.

4. Known Limits and Design Responses

5. Functional Requirements

FR-1: Roster resolution Build the attendee roster from the participant session log at meeting end. Merge in calendar invitees for context, but do not treat invitees who never joined as attendees. Each roster entry is tagged as identified or unidentified.

FR-2: Summary generation Produce a summary containing: - A 3–5 sentence overview. - A decisions list. Each decision has a one-sentence statement, a timestamp, and a transcript link. Decisions are not assigned to individuals unless the transcript clearly states who made them. - An action items list (FR-3).

FR-3: Action item extraction Each action item includes: description, owner (a roster member), due date if stated, timestamp, transcript link, and attribution confidence (high, medium, low).

FR-4: Hard constraint: no non-attendee owners (must-pass) Before any summary is stored or sent, a deterministic validator checks that every action item owner matches a roster entry from FR-1. If an owner does not match, the item is removed and logged as a pipeline error. This check does not depend on the model or on confidence scores. The model is never the final authority on membership.

FR-5: Unidentified owners If the model attributes an action to an unidentified participant, the item is kept, but the owner is shown as "Unidentified participant (Name as joined)." These items are never emailed. They appear in the organizer's review view only.

FR-6: Confidence gating for delivery Each action item is sent automatically only if all of the following are true: - Owner is an identified roster member. - Attribution confidence is high. - The supporting transcript segment has no overlapping speech flag. - The item's source is not a multi-person room device.

Items that fail any check go to the organizer's review queue. The organizer can confirm, reassign to another roster member, or dismiss. Confirmed items are then sent.

FR-7: Multi-person sources When an action item comes from a multi-person source, the owner field is set to "Confirm owner" and the item is held for review. The review UI lets the organizer or the item's likely owner claim it from the roster.

FR-8: Delivery - Auto-sent items go to the owner's email within 5 minutes of meeting end. - Emails contain: the meeting name, the action item, the due date if any, a link to the summary, and a link to the transcript moment. - Each meeting produces one email per owner, grouping that owner's items. - Emails go only to identified roster members. Admins can disable delivery to external email domains.

FR-9: Review queue Organizers see a summary page with held items, can edit or confirm them, and can see which items were auto-sent. Edits made after sending generate a correction email to the owner.

FR-10: Controls and privacy - Admins can turn summaries on or off per workspace and per meeting type. - Any participant can opt out of their own speech being summarized. Their items are excluded from the summary and their name is removed from the roster. - Summaries and transcripts follow the same retention and access rules as the source recording.

6. Sales Requirement: Immediate Auto-Send

Sales wants action items emailed as soon as the meeting ends. This PRD supports that for high-confidence items (FR-6, FR-8). It does not send low-confidence items immediately, because a wrong owner gets a task they didn't agree to, and an email that is later corrected erodes trust. Sales can see the auto-send rate in the admin dashboard. If the held rate is too high for their workflows, the threshold can be tuned with data from the pilot. It should not be lowered without evaluation.

7. Acceptance Criteria

• AC-1: Across a 300-meeting eval set, zero action items name an owner absent from the participant log. This is a hard gate for launch.
• AC-2: Every decision and action item has a working transcript link that lands within 10 seconds of the cited statement.
• AC-3: For a labeled eval set, owner precision on auto-sent items is at least 97%.
• AC-4: Summaries are generated within 5 minutes of meeting end for 95% of meetings with transcripts under 90 minutes.
• AC-5: An uninvited identified joiner appears in the roster and can receive items. An unidentified joiner never receives email.
• AC-6: Opt-out removes the participant's items and name from the output within one processing cycle.

8. Metrics

• Adoption: Percentage of eligible meetings with a summary delivered.
• Quality: Action items edited or reassigned by the organizer, by confidence band.
• Trust: Post-send corrections per 100 auto-sent items. Target: under 2.
• Usage: Percentage of owners who open the linked transcript from an email.

9. Open Questions

1. Should the review queue notify the organizer by email, Slack, or in-app only?
2. When an owner is an external email address, what is the policy under the admin domain setting?
3. How will we handle meetings where the transcript contains no clear owner for a decision or action?
4. Do we need a per-workspace setting to require organizer approval for all items?
5. Should multi-person room devices be fixed at the hardware or SDK level, which would change FR-7?

10. Phasing

• Phase 1 (pilot, 20 workspaces): FR-1 through FR-5, FR-8 through FR-10, with all items held for organizer review. This validates the constraint and attribution quality.
• Phase 2: Enable confidence gating (FR-6) and auto-send for Sales and other pilot workspaces, based on Phase 1 precision data.
• Phase 3: Multi-person source handling at scale (FR-7), and general availability.

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.

PRD: AI Meeting Summaries with Decisions and Action Items

Status: Draft for engineering and design leads Product: Huddle

1. Problem

Huddle records and transcribes meetings with speaker labels, but attendees still write their own notes afterward. Notes are late, inconsistent, or never written, so decisions get lost and action items have no clear owner. We will generate a summary after each meeting that lists decisions and proposes action items with owners.

2. Goals and non-goals

Goals - Produce a summary within 5 minutes of meeting end: decisions, action items (task, owner, due date if stated), and open questions. - Never assign an action item to someone who was not in the meeting (hard constraint, see §4). - Keep the host's review to under 60 seconds for a typical meeting.

Non-goals (v1) - Fully automatic emailing of action items to owners (see §3). - Cross-meeting task tracking or integrations with Jira, Asana, and similar tools. - Summaries for meetings without a transcript.

3. Decision on the Sales request

Sales asked for action items to be emailed automatically to owners the moment a meeting ends. We are not building that for v1. Instead, we will send the draft summary to the host at meeting end. The host can send it to owners with one click, and owners are emailed on send.

Reasons: - Speaker labels are wrong about 8% of the time. An automatic email would sometimes tell a real person they own a task they never agreed to. Because it goes out with no human check, the error becomes visible to a customer's colleagues. - Email is irreversible, so a wrong or non-attendee assignment cannot be quietly fixed afterward. - The underlying Sales need is that owners hear about their tasks quickly and nothing falls through the cracks. A host prompt at meeting end with one-click send meets that need at the cost of a few seconds.

Path to automation: Auto-send will be an opt-in, per-workspace setting, considered after launch once the gates in §9 are met. It would send only high-confidence, presence-verified items and leave everything else for host review. Engineering should build the send pipeline so that auto-send is later a policy change rather than a rebuild.

4. The hard constraint: who can own an action item

Rule: An action item's owner must be a participant who was actually in the meeting when the item was discussed. Anyone else is ineligible.

This rule cannot rely on the calendar invite, and it cannot rely on the model behaving well.

Attendance source. The eligible-owner roster is built from Huddle's session join/leave logs, not from the calendar invite. This handles both failure modes: - Invitees who never joined are not on the roster. - The roughly 15% of meetings with uninvited joiners include those people, matched to a user account when possible.

Presence at time of item. A person is eligible for an item only if they were in the session at the timestamp of the relevant utterance. Someone who left at minute 10 cannot own an item from minute 40.

Enforcement is structural, not prompt-based. 1. The extraction step receives the roster as a list of participant IDs and must return an owner ID from that list or `null`. 2. A deterministic validator runs after the model. It rejects any owner ID not on the roster or not present at the item's timestamp, and it converts that item to unassigned. 3. The UI and email layers accept only validated owner IDs. There is no free-text owner field on the model output path. 4. Names mentioned in the transcript who were not present ("have Dana look at it") produce an unassigned item. The text may keep the speaker's wording, but Dana gets no assignment and no email.

Shared rooms. A room device usually appears as one participant, so individuals in the room are not in the presence log. We cannot verify that a named person is in the room. Such items become unassigned by default, and the host can attest who was in the room during review (§6). That attestation counts as human-verified presence.

Uninvited joiners without an account (guests) can be eligible owners, shown by display name. They receive email only if the host supplies an address during review. We never guess addresses.

5. Handling speaker-label errors

The 8% label error rate is concentrated in crosstalk and shared rooms. Eligible-owner enforcement does not solve it, because a mislabel can still pick the wrong person who was present. Mitigations:

Confidence tiers (High/Medium/Low) are stored on each item, shown to the host in review, and used later to gate auto-send. Thresholds are set empirically during the evaluation in §9. The tiers themselves are a v1 requirement.

6. User experience

Host flow 1. The meeting ends and the host gets a notification: "Summary ready to review." 2. The review screen lists: - Decisions as short statements, each linked to the transcript moment. - Action items with the task, owner chip, due date if stated, and a confidence indicator. - Unassigned items grouped at the top of the actions list, so gaps are visible. 3. The host can edit text, reassign from a picker limited to eligible people, delete items, or add room attendees ("Who was in the room?"). Adding room attendees makes them eligible for items in the room segments. 4. Send emails each owner their own items plus the decisions. Unassigned items go to no one.

Design requirements - Every item has a "jump to source" link that plays the transcript segment. Reviewing should be verification, not re-listening. - The owner picker never offers non-attendees. If the host wants to assign to an absent person, show a separate option, "Send as a follow-up (not from the meeting)," which is a different item type with clearly different wording and no "you agreed to" language. Design to decide whether this ships in v1; the default is that it does not. - Low-confidence owners appear as suggestions ("Suggested: Sam") and never as assignments. - Email copy says "Action item from Meeting name, date", includes the source excerpt, and offers a "This isn't mine" link that notifies the host. This gives owners a correction path. - If the host does not review, send a reminder at 24 hours. Nothing is sent to owners without host action in v1.

7. Functional requirements

1. Roster service: builds the per-meeting roster with join/leave intervals from session logs and links to user accounts where possible.
2. Extraction: from the transcript and roster, produce decisions, action items, and unassigned items. Each action item has task text, owner ID or null, due date or null, source timestamps, and a confidence tier.
3. Validator: enforces §4 as code with unit tests, independent of the model, and logs every rejection.
4. Review UI: as described in §6, on web and desktop. Mobile is view-only for v1.
5. Delivery: per-owner emails on host send, with a record of who was emailed. Sends are idempotent, and edits after sending trigger an explicit "send update" action.
6. Audit log: stores for each item the model output, validator result, host edits, and send events.
7. Admin setting: workspace-level toggle for summaries, plus retention aligned with existing transcript retention.

8. Edge cases to specify and test

• Invitee never joined but is mentioned as owner.
• Uninvited joiner takes an action; the guest has no account.
• Person leaves and rejoins (multiple intervals).
• Two people share a first name; ownership must resolve by roster, and ambiguity means unassigned.
• Whole meeting in a shared room (all items default unassigned until the host attests).
• Multiple languages, or a transcript with large gaps.
• Nobody commits to anything: show "No action items detected" rather than inventing some.

9. Success metrics and launch gates

Launch gates (must all pass) - Zero non-attendee assignments across a labeled evaluation set of at least 500 meetings, deliberately oversampled for uninvited joiners, shared rooms, no-show invitees, and crosstalk. Any production occurrence afterward is a Sev-1 incident. - Among High-confidence items, wrong-attendee assignment rate of 3% or less (target, to be validated against the evaluation set; propose a stricter figure if data supports it). - Median summary delivery within 5 minutes of meeting end.

Post-launch metrics - Share of meetings where the host sends the summary within 1 hour. - Host edit rate per item, and owner "This isn't mine" rate. - Share of action items left unassigned. Some is healthy, but a high rate means the model is too timid.

Auto-send criteria (future): two consecutive months of High-tier precision above the agreed bar and no non-attendee incidents. Sales and Support are consulted before enabling.

10. Risks and open questions

• Sales disagreement. Sales may see the host step as friction. We should share §3 with them early and offer a pilot of auto-send with a few design-partner accounts after the gates are met.
• Hosts skipping review. If sending rates are low, the fix is better reminders and faster review, not removing the check.
• Presence data quality. Do our session logs reliably capture join/leave for all clients and dial-in users? Engineering to confirm in week 1.
• Guest privacy. Can we email or display guest names without consent, given customer policies? Legal to advise.
• Room attestation. Is host attestation adequate for the constraint, or should room participants confirm individually? Recommend host attestation for v1.
• Follow-up items for absent people. Ship in v1 or defer? Design and Product to decide.

11. Milestones

1. Weeks 1–2: roster service, presence data audit, evaluation set labeling begins.
2. Weeks 3–5: extraction, validator, confidence tiers, offline evaluation.
3. Weeks 4–7: review UI and email delivery, in parallel.
4. Weeks 8–9: internal dogfood, gate review.
5. Week 10: limited beta with design-partner customers, including Sales' accounts.