Hi, I’m Aashrith S Dontha
Senior Product Manager | AI, Data & Platforms
Senior Product Manager with 5+ years building AI, data and platform products across regulated enterprise environments and a production SaaS startup.
5+
Years PM
1M+
Users Served
6
Product Lines on Platforms I Own
0
Cutover Incidents Across 2 Platform Retirements
1
Production SaaS Founded

About Me
I'm a Senior Product Manager with 5+ years building AI, data and platform products across regulated environments. I've served 1M+ users, built a data platform from zero at ATB and founded a production SaaS product.
Professional Experience
Building and scaling AI-powered products across banking, telecom, and enterprise platforms.
Product Manager
RBC · Toronto, ON
Own AI, data and API platforms across a six-product-line portfolio, including a 0-to-1 document-intelligence product and a partner-channel review workflow.
Founder & Product Lead (part-time)
Navyashu Applied Sciences & Enterprise · Toronto, ON
Built a production, multi-tenant consumer AI SaaS product with subscription billing, integrations and security controls as part of a two-person team.
Product Owner
TD Bank · Toronto, ON
Owned a two-channel, 12-system decisioning flow for 1M+ customers and built the experimentation and partner-integration practices around the platform.
Senior Product Analyst / Lead BA
Bell Canada · Toronto, ON
Aligned launch readiness across fourteen systems, supported a phased migration of 3M+ accounts and shipped a two-sided field-service product.
Product Analyst
ATB Financial · Calgary, AB
Built a data platform from zero, improving insight turnaround for five teams, and supported an AI virtual-assistant pilot.
Featured Projects
Key projects I’ve led across AI, payments, cloud, and data platforms — delivering measurable business impact at scale.
RBC · Partner-channel AI · Document intelligence
AI Document Review for Partner Channels
Extended a 0-to-1 document-intelligence platform into a high-volume partner channel. The workflow replaced repetitive manual checks with exact-match automation and routed exceptions to people for review.
160+ per case
Manual Checks Replaced
96.35%
Field Accuracy
-60%
Decision Cycle Time
Key Outcomes:
- Replaced more than 160 manual checks per case with exception-only review; exact matches passed and mismatches went to an agent with the reason shown.
- Measured 96.35% field accuracy when a value was returned, against a 95% bar, on a 230-document validation set.
- Designed for 20,000 cases per month and 900 per hour at peak as capacity requirements, not observed traffic.
- The broader AI platform reduced decision cycle time by 60% and saves about 4,200 analyst hours annually across three business lines.
RBC · Senior Product Manager
Central API Platform
Helped replace point-to-point integrations with one reusable API contract for a multi-product lending portfolio, reducing duplicated integration work and making downstream status and errors predictable.
6
Product Lines
7+
Downstream Consumers
~50% faster
Integration Time
Key Outcomes:
- Defined one contract serving six product lines and more than seven downstream consumers.
- Product integration became about 50% faster in the year-end review.
- Standardised authentication, validation, status, error, throttling and versioning conventions.
- Maintained backward compatibility and regression-tested changes across consuming systems.
RBC · Senior Product Manager
Data Foundation
Established the quality gates, lineage and onboarding standards for the data foundation underneath a six-product-line lending portfolio, so product teams could use consistent, validated information.
3
Data Layers
Months → weeks
New Source Onboarding
Shared hub
Product Events
Key Outcomes:
- Introduced staging, canonical and production layers with a quality gate between each.
- Reduced new data-source onboarding from months to weeks with reusable patterns.
- Added lineage for compliance and service levels with named owners.
- Streamed product events into one hub so funnels could be measured consistently.
RBC · Senior Product Manager
Embedded Checkout Partnership
Launched an embedded financing product with a national retailer, defining pricing, decision rules and a measurable activation journey where no internal product baseline existed.
0 → live
Product Status
~30% faster
Time to Payout
3 days
Activation Window
Key Outcomes:
- Launched a financing product at a national retailer, displacing an incumbent instalment provider.
- Set activation within the three-day virtual-card window as the North Star, with delinquency as a guardrail.
- Time to payout was about 30% faster than the displaced process.
- Published canonical business events through a reusable pattern adopted by later lending products.
RBC · Senior Product Manager
AI Documentation Assistant
Built repository-reading agents that produced evidence-backed service wikis, then proposed a citation-first knowledge assistant for employees based on those structured sources.
512 → 37
Repositories → Wikis
~40% faster
Engineer Onboarding
~30% fewer
Support Questions
Key Outcomes:
- Processed about 512 repositories into 37 structured service wikis.
- Teams estimated engineer onboarding was about 40% faster and questions to senior engineers about 30% fewer.
- Required every generated claim to cite a file path or be marked UNKNOWN.
- Secured director sponsorship for a read-only employee knowledge-assistant proposal.
TD · Product Owner
Experimentation Practice
Created a rigorous experimentation practice on a real-time decisioning platform, replacing unverified “wins” with pre-registered tests, holdouts and measurable outcomes.
1M+
Users Served
+18%
Engagement
-12%
Churn
Key Outcomes:
- Built pre-registration, holdout and significance standards for experiments.
- Ran 15+ experiments; engagement improved 18%, conversion 18% and churn fell 12%.
- The experimentation framework was adopted by three other product teams.
- Redirected engineering from a low-value data-enrichment initiative, avoiding about $2M in annual spend.
TD · Product Owner
Partner Integration Platform
Turned a partner integration journey into a measurable funnel, discovering that sandbox provisioning—not documentation—was the largest onboarding bottleneck.
3 weeks → <1
Integration Time
$1.2M
New Revenue
3×
API Throughput
Key Outcomes:
- Reduced partner onboarding from three weeks to under one week.
- Partner integrations grew 40%.
- Generated $1.2M in new integration revenue and improved API throughput threefold.
- Automated sandbox provisioning and avoided an unnecessary documentation rewrite.
TD · Multi-channel platform, regulatory change, sunset
Two Channels, Twelve Systems, 120 Seconds — Then Three Regulatory Changes on the Live Flow
A vendor delivered a skeleton decision engine. We configured it into live production for a million customers across two channels — a national e-commerce partner with a two-minute decision promise and the direct channels — redesigned the API architecture around it, absorbed three regulatory changes on the live flow, and retired the on-premise engine it replaced.
1M+
Customers
120 s / <500 ms
Partner SLA / Decision P99
0
Cutover Incidents
Key Outcomes:
- Automated approval rate increased 22% after third-party behavioural signals were integrated.
- Held the 120-second partner SLA and sub-500ms decision API P99 for 1M+ customers.
- Delivered three regulatory changes on the live flow with zero compliance exceptions in post-launch audit.
- Decommissioned the legacy engine with zero cutover incidents; Definition of Ready reduced rework 30%.
Bell · Senior Product Analyst / Lead BA
Self-Serve App from 500 Verbatims
Read more than 500 raw customer comments and found that support dependency—not product quality—was driving dissatisfaction. Redirected an already-planned quality sprint toward the self-serve actions customers needed.
+14 points
NPS
-25%
Support Volume
+10%
Upgrade Conversion
Key Outcomes:
- Read 500+ verbatims and identified forced support contact as the underlying source of customer friction.
- Improved NPS by 14 points and reduced inbound support volume by 25%.
- Usage-triggered upgrade prompts lifted conversion 10% with zero acquisition spend.
- Retired a lower-performing legacy flow without complaints or re-engagement requests.
Bell · Launch readiness, migration, two-sided product
Fourteen Systems to One Launch Date, 3M+ Accounts Moved, and a Field App Dispatchers Could Act On
Three infrastructure programmes in one year: a network-product launch that fourteen downstream systems had to be ready for, a phased migration of 3M+ accounts to a new platform, and a two-sided web and Android field-service product for technicians on company-provisioned devices.
14
Systems Aligned
3M+ / 99.9%
Accounts / Uptime
-40%
Dispatcher Response on Exceptions
Key Outcomes:
- Held the launch date after a readiness assessment surfaced a billing gap six weeks ahead of launch.
- Migrated 3M+ accounts in phases with rollback criteria, 99.9% uptime and no service disruption; validated 2× capacity.
- Cut dispatcher response time on exception cases by 40% with six specific exception codes and automatic rerouting.
ATB · Product Analyst
Data Platform from Zero and AI Pilot
Built a data lake, event schema and self-serve dashboards from zero, then used the resulting data to find a validation defect behind funnel drop-off and support a measured AI virtual-assistant pilot.
5 days → <4 hours
Insight Turnaround
5
Teams Unblocked
+8%
Conversion Lift
Key Outcomes:
- Built a data lake, event schema and dashboards in three months for five cross-functional teams.
- Reduced insight turnaround from five days to under four hours.
- Found a data-validation edge case; fixing the logic lifted conversion 8% and reduced time-to-activation 20%.
- The ATB virtual-assistant pilot achieved 20% case deflection, worth about $150K annually.
Navyashu · Founder & Product Lead
Founder-Built Consumer AI Product
Built a production, multi-tenant consumer AI product with subscription billing, integrations and explicit security controls as a two-person team with near-zero budget.
11 · 23
Integrations · Custom MCP Tools
17 → 0
Security Findings
59 / 59
CI Tests Passing
Key Outcomes:
- Shipped subscription billing with three metered tiers and migrations for existing users.
- Built eleven production integrations and an agent layer across more than ten channels.
- Closed 17 security findings through three versioned migrations with zero downtime.
- Maintained tenant isolation and passed all 59 CI tests; no user, revenue or MRR figures are claimed.
Skills & Tools
Technologies and tools I use daily to build high-quality products.
Product & Strategy
AI/ML & Data
Growth & UX
Tools & Platforms
Latest from the Blog
Thoughts on web development, design, and building products that matter.
Your API’s Error Contract Is Part of the Product
A shared API is not just a successful request shape. Predictable errors, status, versioning, and recovery determine whether downstream teams can build and operate with confidence.
Retire a Legacy Platform Without Making Cutover the Experiment
A safe sunset is designed before traffic moves: define comparison criteria, shadow the replacement, agree rollback thresholds, and make the operating model part of launch readiness.
The Safest AI Automation Knows When to Stop
For high-stakes document review, confidence scores were not enough to auto-accept a value. An exception-only design automated exact matches and made uncertainty visible.
Let’s Build Something Amazing Together
I’m always open to discussing new opportunities, exciting projects, or just connecting with fellow professionals. Let’s start a conversation!