Built the core Compose Skill in the NestJS AI service. The skill generates personalized outreach messages for brands contacting creators. Architected a multi-model pipeline: Claude handles instruction parsing (extracting brand requirements from natural language), OpenAI handles email composition (generating natural-sounding outreach copy). Models orchestrated through Databricks. Added retry logic with error recovery, enhanced JSON parsing for malformed inputs, and placeholder resolution for unresolved content variables.

Impact: Brands go from 'I need fitness creators in LA with 50K+ followers' to personalized outreach emails in a single chat conversation, replacing a manual process that took hours per campaign.

Why this approach: Multi-model approach chosen because instruction parsing and creative writing are fundamentally different tasks. Claude excelled at structured extraction, OpenAI at natural-sounding copy. Databricks provided ML orchestration. Retry logic with graceful degradation because outreach emails are high-stakes (a broken email harms the brand's reputation).

Built the Affiliate Hub frontend (Next.js 15 / React 19) and its platform embedding. Implemented collapsible chat sidebar, real-time message count streaming via Server-Sent Events, creator approval workflows with bulk actions, CRM profile linking, and agent settings management. On the backend, built the SSE endpoint with MongoDB change streams optimized by account ID filtering. On the main platform (PHP/Vue, 160 commits), embedded the Affiliate Hub via iframe with feature flags and routing.

Impact: The Affiliate Hub is a critical revenue-generating feature (messaging creators to start partnerships). Real-time streaming means campaign managers see message delivery stats live instead of refreshing the page. Bulk approval workflows let teams process creator lists in seconds instead of one by one.

Hardened internal API endpoints by adding organization scope enforcement in the resource manager. Refactored AffiliateContactLinkMiddleware with proper filtering and deduplication to prevent data leakage between organizations. Secured internal endpoints to prevent unauthorized cross-organization data access.

Impact: Prevented unauthorized data access between organizations on a multi-tenant platform serving thousands of brands. In a B2B SaaS platform, a data leak between customers is a company-ending event.

Built YouTube and TikTok profile insights retrieval and data transformers in the Network Services repo (68 commits). Created standardized ProfileInsightsTransformers for cross-platform data normalization. Added caching providers to reduce API calls, implemented request deduplication, and added brand affinity metrics with audience segmentation. Wrote comprehensive unit tests for all new transformers and providers.

Impact: Expanded the platform's analytics coverage to YouTube and TikTok, giving brands deeper creator insights for campaign decisions. Caching reduced external API load and improved response times for creator profile views.

Integrated Klaviyo for scalable email delivery and Bouncer for real-time email verification. Built the pipeline so outreach emails are verified before send, with CRM contact existence checks to prevent outreach to invalid or already-contacted creators. Added Threads platform integration for multi-channel outreach.

Impact: Brands send outreach at scale without risking their email domain reputation. Contact verification prevents embarrassing duplicate outreach to creators already in the CRM.

Designed and implemented a rules engine that governs transaction limits based on user suitability profiles. Users complete KYC questionnaires (accredited investor status, risk tolerance, investment experience), and the engine enforces per-user, per-jurisdiction spending limits in real-time across all trading operations (buy, sell, deposit, withdraw).

Impact: Brought the platform into compliance with CSA/OSC suitability requirements, unblocking continued operation in Canadian markets. Without this, Netcoins risked losing its securities registration. Previously, transaction limits were either absent or manually enforced by operations staff.

Why this approach: Rule-based engine with questionnaire-driven profiles rather than hardcoded tiers, so new regulatory requirements could be added via configuration without code deployments. This was critical because regulators update requirements frequently and compliance delays carry real legal risk.

Re-architected how the platform determines which assets are available for which operations (buy, sell, deposit, withdraw, stake) in which jurisdictions. Replaced scattered hardcoded checks across middleware, policies, verification rules, and services with a centralized bitmask-based permission system in the database. Each bit represents an allowed operation, making permission checks a single bitwise operation. Integrated with custody providers (BitGo, Fireblocks) with automatic verification that at least two providers support the asset before trading opens (Canadian regulatory requirement).

Impact: Reduced new asset listing time from 2–3 days of code changes across multiple services to a database configuration change that takes minutes. Operations team could manage listings without engineering involvement. This was the single largest refactoring effort during my time at Netcoins.

Why this approach: Bitmask approach chosen for performance (single integer comparison per check) and expressiveness (any combination of operations representable). Centralized in DB so non-engineers could manage listings. The alternative was a config file approach, but that still required deployments and couldn't be audited by compliance teams directly.

Built a complete online enrollment system from scratch in weeks when COVID-19 made paper-based international enrollment impossible. Laravel + MySQL application with AWS S3 document upload. Students fill forms, upload documents, select courses and groups. Staff verify data rather than re-enter it. Integrated with the main Hanson App and synced to external systems (Schoology) via Python scripts.

Impact: 1,400+ international students enrolled through the platform. $20M+ CAD in accepted students that year, a record for the institution. Replaced a process where agents went back and forth via email/phone, received paper PDFs, and manually retyped everything into multiple systems. I was sometimes debugging browser issues at 11pm with students calling from India.

Why this approach: Built as a separate application that integrated with the main system rather than bolting onto it, because speed of delivery mattered more than architectural purity during a pandemic. We started accepting students with a beta version because the alternative was not accepting them at all.

Built the central web application (Laravel + Vue) that served as the operational backbone for the entire college. Announcements, schedules, attendance tracking, grades, document requests. Admin panels for student services, academic department (scheduling), enrollment department (pipeline and status tracking), and VP-level overview dashboards.

Impact: Replaced disconnected manual processes across 4+ departments with a single system. Gave leadership real-time visibility into enrollment pipeline, attendance, and operations for the first time.

Built Android (Java) and iOS (Swift) mobile applications for students from scratch. Features: announcements, class schedules, document requests, event ticket booking, attendance marking. Content managed through the main Hanson App admin panel.

Impact: Students accessed college services from their phones instead of visiting offices or checking email. Reduced foot traffic to student services and improved communication reach.

Built an Android kiosk application installed in student services offices for semester registration. Students enter student ID, verify identity via email code, and select their next semester group. Multi-language support with contextual hints designed for international students so they could complete registration without asking a representative.

Impact: Reduced student services department workload by approximately 50% during registration periods. International students could self-serve in their preferred language instead of queuing for a representative.

Mentored 5 cohorts of approximately 15 students each (75 total) on capstone projects. Scoped feasible 3-month projects, helped teams split work, ran progress syncs, and ensured delivery before semester end.

Impact: All cohorts delivered working projects on time. Students gained practical experience in project scoping, teamwork, and delivery under deadline.