Kindo
Kindo × Deloitte Program
Program Meeting · August 4, 2026

Swimlane Replacement.
Turbo Mode Walkthrough

Joint session to align on Kindo's Turbo Mode solution and Deloitte's current Swimlane usage. Goal: walk out with a clear understanding of what needs to be done, a phased rollout strategy, and all open questions answered.

Session Goals
1
Turbo Mode Walkthrough — Kindo walks Zun through the current solution developed by the Kindo team
2
Swimlane Current State — Zun walks us through how Deloitte is using Swimlane today, including the capabilities needed from the new solution
3
Alignment & Next Steps — Establish clear understanding, phased rollout strategy, and resolve all open questions
Attendees
Tony Wong · Charlie Hulcher · Joana Dias · Victor Slompo · Zun Huang (Deloitte)
Swimlane Replacement — Journey So Far

Tracking every step of the Swimlane replacement workstream. Swimlane contract ends February 2027, with a go/no-go decision by December 2026.

July 21, 2026
Kindo × Deloitte — Program Session Biweekly
First biweekly program call. Platform improvements discussed — observability, MCP latency, flow control. Swimlane replacement surfaced as a cross-cutting topic.
July 27, 2026
Swimlane Discovery — Walkthrough Session
Deep-dive into current Swimlane usage. Understanding workflows, integrations, and operational dependencies that need to be addressed in the replacement.
July 29, 2026
Material Received from Zun + Charlie's Review
Zun shared capability gap requirements document. Charlie reviewed and provided detailed assessment — categorizing each requirement by platform readiness and Turbo Mode impact. This analysis frames today's session.
July 30, 2026
Kindo/Deloitte Agent Design Work Session
Design Session #2 — platform readiness, observability, Swimlane transition direction, and SOC for AI co-design. Zun shared 10 capability gap requirements. Turbo Mode demo planned.
August 4, 2026
Program Meeting — Turbo Mode Walkthrough & Swimlane Deep Dive
Today's session. Kindo demonstrates Turbo Mode to Zun. Zun walks through current Swimlane usage and required capabilities. Goal: phased rollout strategy and all questions answered.
Agenda
Block A — Turbo Mode Walkthrough (Kindo → Deloitte) · 25 MIN
Turbo Mode was demonstrated during the July 30 Design Session. This block provides a walkthrough for attendees who were not present, followed by an open discussion on how it maps to current Deloitte workflows.
Turbo Mode — Walkthrough & Discussion
Review the Turbo Mode product tour from the July 30 session. After the walkthrough: where does this map to current Swimlane workflows? Where are the gaps? What needs further design?
Capability Mapping — Turbo Mode vs. Zun's Requirements
Map Turbo Mode against the 10 SOAR capability requirements Zun shared July 29 — identify what's impacted, what's in progress, and what needs further design.
Block B — Swimlane Current State (Deloitte → Kindo) · 25 MIN
Zun walks us through how Deloitte is using Swimlane today — focusing on what we don't yet know. The July 27 Discovery session covered the topics below. This session should go deeper into gaps and specifics not yet captured.
What we already know from July 27 Discovery:
  • Alert injection — Swimlane is the single point of entry for all client alerts (email-based + API-based). 80% of clients use it.
  • Pre-processing pipeline — Normalization, deduplication, correlation before creating Jira tickets
  • Current Kindo integration — Swimlane already triggers the triage agent in Kindo production
  • Playbooks — Hundreds of individual playbooks per use case. Playbooks call other playbooks (cross-communication). Client-facing (sold as customized) + internal playbooks.
  • Deterministic flow — Drag-and-drop action blocks with conditions + custom Python code. If/then/else logic that agents can't reliably handle today.
  • Token cost concern — Don't want to burn tokens for basic API work that doesn't need LLM
  • Multi-tenancy — Main tenant managing multiple client tenants with versioning
  • Playbook maintenance — Engineers (Harshal, Archith) constantly monitor and fix when upstream data changes
  • Requirements doc — 10 categories, 7 pages, prioritized, created April 2026
Today's focus — what we still need:
  • Specific playbook examples — Zun offered to share sample playbooks. Have they been shared? Can we walk through 2–3 representative ones?
  • Production Swimlane walkthrough — Zun mentioned needing Krishna/Kush approval to show production. Has that been granted?
  • MSSP orchestration details — How exactly does multi-tenant playbook management work? Versioning, customization per client?
  • Parallel flows and cross-agent communication — Archith raised parallel execution and conditional branching. What does this look like in practice?
  • Volume and load specifics — Numbers on alert volume, concurrent playbook executions, peak load
  • Anything beyond the 10 SOAR capability requirements already shared on July 29
  • Migration input format — Krishna asked (July 30) whether the starting point always has to be an agent, since they already have explicit logic. Charlie mentioned possible migration tooling. Across ~150 workflows, pasting logic one at a time vs. having tooling is the difference between feasible and not.
  • MCP coverage — Adelina flagged that they use webhooks and direct APIs, and some technologies may not have an MCP server. Turbo Mode runs on Kindo's MCP tools. Blocking dependency, unmapped.
  • Low-code requirement — Adelina said this would likely be a requirement. Effectively an eleventh item if real.
What We Don't Know Yet — Swimlane Deep Dive
Beyond what was covered in the July 27 Discovery session: operational dependencies, edge cases, analyst workflows, and integration specifics that haven't been surfaced.
Required Capabilities Beyond the 10 Gaps
Are there requirements not captured in Zun's July 29 document? Non-negotiable capabilities, operational constraints, or compliance requirements the new solution must meet.
Gap Analysis — Swimlane vs. Kindo + Turbo Mode
Given the Turbo Mode demo in Block A and the capability gap assessment: where do real gaps remain? What needs engineering investment vs. configuration vs. workflow redesign?
Block C — Alignment & Phased Rollout Strategy · 10 MIN
Zun's own requirements document concludes that closing the gaps will require vendor enhancements, complementary tooling, "or a phased rollout that limits exposure while gaps are addressed." Phased rollout is Deloitte's recommendation — this block defines what that looks like.
Phased Rollout Strategy
Define a phased approach: what can be transitioned first, what requires more development, and what the migration timeline looks like against the February 2027 contract end.
Open Questions — Resolution
Address all outstanding questions from both sides. Every open question gets an explicit owner and date for resolution.
Next Steps & Owners
Named actions, named owners, dates. What happens between now and the next checkpoint.
SOAR Capabilities We Need Zun's Requirements · July 29 + Kindo Assessment

Zun shared 10 SOAR capability requirements on July 29. Krishna directed this be framed as "SOAR capabilities that we need" with a roadmap question attached. Charlie reviewed each one and assessed platform readiness. Below is the categorized assessment — framing today's discussion.

Would Be Impacted by Turbo Mode 5 OF 10

These requirements would be directly impacted by Turbo Mode, if Kindo ships it.

1.
Alert Intake & Scale — Alert ingestion is solely agentic; no way to handle large volumes without using LLM.
2.
Data Normalization — No standard schema mapping / normalization across sources, adding cost and inconsistency to downstream processing.
3.
Noise Reduction — No deduplication/suppression layer; repetitive alerts drive avoidable processing cost.
4.
Incident Correlation — No native way to group related alerts into a single incident; each run is isolated.
5.
Workflow Control — Workflows are reasoning-based only, with no deterministic layer for actions requiring guaranteed, repeatable execution.
Under Active Consideration IN PROGRESS

Projects under consideration that would address this requirement. Not yet implemented.

6.
Response Actions — No pre-built action library with structured approvals, rollback, or safeguards against runaway automation.
Requires Strategic Discussion DISCUSSION

These items need further alignment between Kindo and Deloitte on approach and scope.

7.
Case Management — No persistent incident lifecycle or system of record; reliant on external ticketing tools. Kindo's platform direction favors AI-native approaches over traditional systems of record — this is a key discussion point.
Dependency note: Items 3 (Noise Reduction), 4 (Incident Correlation), and 9 (Reporting & Metrics) all depend on this. Deduplication needs to query prior incidents, correlation needs an incident object, reporting needs records to query. If #7 goes unanswered, four of ten go unanswered.
8.
Multi-Tenancy — Client isolation and per-client configuration cited as limited. Kindo is designed for multi-tenancy — need to understand the specific gaps Deloitte is experiencing to address them.
Addressable — Scope Needed SCOPING

Potentially low-lift to address once specific requirements are defined by Deloitte.

9.
Reporting & Metrics — Operational, SLA, and performance dashboards fall short of SOC reporting standards. Need Deloitte to define what "SOC reporting standards" means specifically so Kindo can scope the work.
Known Limitation NOT YET PRIORITIZED

Known platform limitation — not currently prioritized within Kindo's roadmap.

10.
Secure / Edge Deployment — No lightweight on-prem or air-gapped execution option. If client tools are behind a firewall and Kindo is deployed as SaaS, PrivateLink or site-to-site VPN is required.
Critical Timeline
November 2026 — Migration must begin. Zun: "We can't wait until December and then have one or two months to migrate all our clients to new systems."
December 2026 — Go/no-go decision. Swimlane requires 60 days notice before contract end. This is the hard deadline for renewal or cancellation.
February 2027 — Swimlane contract ends. No fallback.
Source: Zun Huang, July 27 Discovery Session. Tony and Mo committed to this timeline at RSA (March 2026) with Kush and Krishna present.
Required Outcomes from This Session
  • Clear understanding of what needs to be done
  • Phased rollout strategy with milestones
  • All open questions answered by Deloitte — or assigned an owner and date
  • Gap analysis: Turbo Mode vs. Swimlane capabilities
  • Named owners for each migration workstream
Kindo

Kindo × Deloitte — Program Meeting · August 4, 2026