Projects

GTM decision velocity platform — CRM alignment, medallion architecture, and near-real-time reporting workflows.

Digital backbone for public schools in Zimbabwe/Zambia — an operational nervous system, not 'just another school app'.

Incident intelligence system inside Microsoft Teams — turning operational chaos into structured decision objects.

Operational Decision Infrastructure — treating incidents, alerts, tickets, and risk signals as 'decision objects' in a nervous system.

Signal-driven portfolio intelligence — risk scoring, benchmark comparisons, and regime-aware insights.

Agentic segment drift and reach intelligence — continuous monitoring with intelligent explanations when audiences degrade.

AI-powered SaaS tool recommendation platform — constraint matching, risk assessment, and decision brief generation.

Subscription-first urban mobility — predictable driver earnings, reliable rider experience, and safety-scored infrastructure.

For distributed engineering teams struggling with integration chaos, Aegis ensures APIs stay secure, compliant, and contract-safe. Implements schema validation, policy enforcement, and cross-service reliability checks across ecosystems.

Local multimodal pipeline (vision → LLM → RAG) with an enterprise privacy posture: local-first, optional hybrid routing, benchmarks, and observability.

For enterprises drowning in alerts and fragmented tooling, Agent Guard becomes the AI control layer that triages incidents, routes decisions, and triggers safe auto-remediation. Built on multi-agent orchestration, policy-as-code, and contextual memory to reduce operational decision time from hours to minutes.

CI/CD quality gate for AI agents. Validates correctness, safety, determinism, and compliance before deployment.

AI Product Strategy Lab for Financial Institutions Design, evaluate, and launch AI products for banking. Work through a structured strategy lab that produces defensible product definitions, differentiation strategies, and go-to-market plans.

Enterprise-grade platform evaluation for GenAI workloads 5 evaluation engines analyse workload alignment, security posture, vendor lock-in, migration complexity, and capability coverage. Make confident, traceable platform decisions.

Is your bank ready for enterprise AI? Evaluate AI maturity, model deployment risks, and generate a strategic adoption roadmap for financial institutions.

Enterprise planning tool for LLM-powered applications. Estimate infrastructure, model, and retrieval costs before deploying AI systems Compare models, optimise spend, and make informed infrastructure decisions — before you write a single line of code.

For cloud-native teams firefighting outages and blind spots, Infra Guard proactively detects drift, latency spikes, and hidden fragility. Ingests observability streams, maps service dependencies, and applies AI-assisted root cause inference to protect uptime before customers are impacted.

Personalized Multi-Agent LLM Trading Recommendation Engine MyTrade deploys a team of specialized AI agents — analysts, researchers, a trader, risk managers, and a portfolio manager — to collaboratively evaluate market conditions and produce actionable trading recommendations

GitHub Profile README

For card issuers seeking higher engagement, Pixel-Pulse identifies behavioral signals that trigger smarter rewards and interventions. Processes high-volume transaction data with segmentation models and real-time decisioning logic.

For leaders overwhelmed by inbox chaos, Pluto converts emails into prioritized decisions, summaries, and tracked actions. Combines LLM reasoning, contextual embeddings, and workflow automation to eliminate cognitive overload and missed follow-ups.

For product and engineering teams struggling to turn raw commits into clear customer communication, Release Notes Composer auto-generates structured, audience-ready updates. Built using commit diff parsing, semantic clustering, and LLM summarization to convert technical changes into meaningful business narratives.

For individuals seeking disciplined wealth growth without constant manual tracking, Robo Advisor builds, monitors, and rebalances portfolios intelligently. Combines asset allocation models, risk profiling engines, and scenario simulations to optimize returns while managing downside exposure.

For compliance-heavy industries, Senet monitors actions in real time and prevents risky decisions before they happen. Implements policy engines, audit trails, and AI reasoning to enforce guardrails without slowing teams down.

Fully local, enterprise-grade multimodal inference pipeline. Images & video → Vision Model → LLM → RAG — zero data leaves your infrastructure.

Define an agent once. Compile to any runtime. Stop guessing what your agents are doing in production. Cortex Protocol wraps any agent - in any framework - with policy enforcement, an immutable audit trail, and compliance reporting that satisfies your auditors.

AI Infra Decision Suite. Generate production-grade AI architectures on Google Cloud Enter your use case, scale, and requirements — get a complete Vertex AI architecture with diagrams, blueprints, cost estimates, and platform rationale.