Ambar
Meridian Capital Partners · Codex Adoption
0%
Agentic Change Management
Change doesn't
wait for humans.
11 specialist agents. One brief. Autonomous from day one.
All 11 agents initialise in 0.3 seconds
Step 1 — Task Assigned
0.3s
Deployment latency
100%
Brief alignment
11
Active specialist agents
Learner Profile Agent — Personalisation at scale
Step 2 — Profile Generation
JC
James Chen
Senior BA · Equities Research
Content sequence generated
vs
SK
Sarah Kim
Junior BA · Operations
Content sequence generated
140 personalised emails sent in 8 seconds
Step 3 — Onboarding Comms
Ambar → priya.sharma@meridian.com
Your Codex onboarding,
Priya
Practitioner track. Risk Analytics workflows: pandas aggregation, SQL window functions, factor visualisation. Training: Wed 16 Apr 14:00.
Practitioner
Risk Analytics
View onboarding deck →
Sent ✓
47
sent
Ambar → james.chen@meridian.com
Your Codex onboarding,
James
Advanced track. Multi-step pipelines, equity attribution, factor analysis. Groupby module skipped — proficiency detected. Training: Tue 15 Apr 09:00.
Advanced
Equities Research
View onboarding deck →
Sent ✓
112
sent
Ambar → sarah.kim@meridian.com
Your Codex onboarding,
Sarah
Foundation track. NAV reconciliation, data quality, report automation for Operations. Excel familiarity noted — bridge modules included. Training: Mon 14 Apr 10:00.
Foundation
Operations
View onboarding deck →
Sent ✓
140
sent
140
personalised emails ·
8
seconds · Compliance-reviewed · All logged
Continuous daily scoring. Every signal captured.
Step 4 — Tracking & Check-ins
Equities Research (28)
Fixed Income (32)
Risk Analytics (40)
Operations (40)
0%
Active adoption
Day 1
Programme day
4.2h
Time-to-insight
0
Flagged
Detects friction in seconds. Not next quarter.
Knowledge Agent — Continuous Monitoring
J. Chen — Codex session · Day 21
⚠ Stall detected — 6 min inactivity · Same prompt sent 3×
11×
Iterations before Knowledge Agent
→
2×
Iterations after resource surfaced
Knowledge Agent · responding
Resources surfaced to J. Chen via Slack DM
Orchestrator detects. Re-plans. Acts.
Real-time Autonomous Re-planning
Fixed Income adoption falls below threshold. 2 seconds to full re-plan. No human involved.
TRIGGER
Fixed Income drops to 8%
31 of 32 BAs below 30% adoption threshold. Cohort at risk.
Day 28 · 09:14:22
→
DETECT
Orchestrator flags risk
KPI breach: adoption target 30%, actual 8%. Programme state updated.
Day 28 · 09:14:23
→
ANALYSE
Root cause identified
Learner Profile: 78% friction on pandas groupby. Training gap confirmed.
Day 28 · 09:14:23
→
RE-PLAN
4 actions dispatched
All agents re-tasked. Running autonomously.
Day 28 · 09:14:24
📚
Training Agent
Generate targeted groupby micro-module for Fixed Income cohort
Re-tasked instantly
🎯
Nudge Agent
Increase check-in frequency: daily instead of weekly for Fixed Income
Re-tasked instantly
🤝
Onboarding Agent
Assign additional champion to Fixed Income from existing cohort volunteers
Re-tasked instantly
📊
Reporting Agent
Flag Fixed Income risk in next SteerCo report with re-plan summary
Re-tasked instantly
2s
Detection to full re-plan
4
Agents re-tasked simultaneously
0
Human interventions required
+29%
Fixed Income adoption after re-plan
Low adoption. Coordination. Human routing.
Step 5 — Intervention
Adoption Agent
J. Chen — 12% index
14 days no sessions. Cohort benchmark 58%. Active Resistant.
Active Resistant
Learner Profile
Friction pinpointed
78% abandonment on groupby. Stalled since Day 21. Velocity: 0.
Gap: pandas groupby
Training Agent
Micro-module ready
20-min module built from J. Chen's actual script patterns.
Content ready
Nudge Agent
Targeted outreach × 2
DM sent with module link. Resistance-calibrated. No response 96h.
No response
Knowledge Agent
Resources surfaced
3 library entries via Slack DM: groupby, multi-index, schema joins.
Delivered
Governance Agent
Comms monitored
All outreach within approved templates. Policy compliance: 100%.
Compliant
✋ Escalation Agent — Human routing
Routed to
Programme Manager
· SLA: 4 hours · "J. Chen — Friction: pandas groupby — Recommend: champion conversation."
No further automated action.
Auto-generated from live agent data.
Step 6 — Report to Management
Month 2 · Auto-generated from live agent data · No consultant required
Meridian Capital Partners — Codex Adoption Dashboard
Month 2 · Reporting Agent · 28 April 2026 · Auto-generated
Pending review
0%
Adoption
4.2h
Time-to-insight
0%
Training done
—
NPS
Adoption by division
Adoption curve week 1→8
Auto-generated · 0 compliance flags · 2 risk flags
Approve & send →
All targets exceeded.
Results — Month 6
68%
Active adoption · Month 6 · Target was 55%
2.1h
Time-to-insight
87%
Training complete
+11
Analyst NPS
40%
Cost reduction
Traditional consulting
6 weeks to assign 140 tracks
One generic onboarding email
Quarterly survey · no real-time signal
3 days building the report
Reactive — wait for escalation
vs
Ambar
Tracks assigned in seconds
140 personalised emails · 8 seconds
Continuous per-BA scoring · instant friction detection
Auto-generated · PM review 10 minutes
Proactive re-planning before failure
Ambar
Agentic Change Management
ambarlab.ai