Selected work.

Early-stage venture needed to ship a GTM motion fast: company registration, partnership funnel, and a 2025–26 roadmap aligned to KPIs — without burning runway.

• Streamlined company registration & policy implementation for 100% compliance.
• Engineered a partnership acquisition funnel with systematic lead generation.
• Led data-driven promotional campaigns to optimise reach and engagement.
• Co-authored the 2025–26 roadmap, aligning ops to long-term KPIs.

The floor was busy, but output and on-time delivery were inconsistent. Supervisors saw utilization, not where time disappeared between cutting, stitching, sleeves, embroidery, ironing, and packaging—or why the same steps varied sharply across operators.

• Introduced lightweight, real-time time tracking per unit via mobile, aligned to the full production chain.
• Made throughput, bottlenecks, and micro-delays between tasks visible to the whole team—not only management.
• Used the data to rebalance work and reduce idle, untracked time instead of adding capacity or hires.

Sellers were losing hours on manual payment ops and burning ad spend on unfocused campaigns. We needed to compress both.

• Integrated Engaze.ai-powered payment systems across the seller network.
• Built Midjourney + Zapier marketing pipelines to auto-generate creative.
• Instrumented funnel analytics to attribute CAC to creative variants.

Distributed environments suffered from unsecured routers and slow incident response — risking data exposure and SLA breaches.

• Upgraded and secured network routers across distributed sites.
• Built automated diagnostic tests for network configurations.
• Documented MTTR playbooks to standardise incident response.

Security professionals needed smarter, automated testing tools for home and office networks. Manual vulnerability scanning is slow, inconsistent, and produces raw data that takes hours to interpret. Teams lacked a way to run AI-directed tests and receive structured, severity-ranked remediation guidance in a single workflow.

• Collaborated on an open-source toolkit integrating AutoGPT (GPT-3 via Langchain) to autonomously run a series of network and system vulnerability tests with no manual orchestration.
• Engineered prompt chains optimised for security-focused Linux environments, enabling the agent to adapt its test sequence based on discovered network topology.
• Built a report generation layer that classifies each detected vulnerability by severity and produces concrete remediation recommendations — turning raw scan output into an actionable security brief.

E-commerce storefronts serve one static experience to visitors with radically different intent. Conversion rates stall at 2–3% industry-wide because brands can't afford the engineering overhead of true personalization. Most leave 30–60% of potential revenue on the table — not from lack of traffic, but from lack of relevance.

• Built a client-side intent engine that reads URL signals (UTM params, referrer, campaign tags) and scores each visitor against 7 behavioral personas in under 50ms — no backend, no deploy, no ops.
• Dynamically swaps hero image, headline, CTA copy, and section order on arrival, creating a tailored storefront experience from the first pixel rendered.
• Added real-time frustration detection — rage clicks, scroll hesitation, exit intent — to trigger recovery flows before bounce, recovering sessions that would otherwise be lost.