"If you've hired developers before and still can't say what actually shipped,
you already know the problem."
I'm a Fractional / Interim CTO who has spent six years being the person founders call to own a build end to end: architecture, development, and deployment, not just execution on someone else's plan.
I work best with early-stage and growth-stage B2B SaaS founders: teams with an idea, an early product, or an AI-built MVP that has outgrown its foundation, and no in-house technical leadership yet to own the outcome.
My deepest, most repeated work is in healthtech and proptech, two verticals where regulatory complexity and legacy-system integration make "just hire a contractor" a genuinely bad bet. I have also shipped in govtech, edtech, and climatetech, but healthtech and proptech are where I specialize.
That ownership model has taken 23 SaaS products from raw idea to live, paying software across 67 client engagements. One recent healthtech client went from a shaky AI-built MVP to a stable, paying product in about three months, using precise LLM context engineering and robust system architecture to turn unpredictable AI into reliable software.
What makes this different from hiring a contractor: I stay past launch. Most of what breaks a product happens after version one, not before, and I am still there when it does.
Where I am not useful: if you already have technical leadership you trust, I cannot fix a strategy problem by writing better code.
- Full technical ownership, not just execution on an existing plan. Architecture, development, and deployment under one roof.
- Post-launch commitment. Staying engaged when most product failures actually happen, instead of exiting after version one.
- AI-to-production hardening. Taking Lovable, Bolt, and Cursor-style "vibe-coded" MVPs and making them secure, stable, and scalable.
- Regulated and complex domains. Healthtech and proptech experience with legacy-system integration and compliance awareness.
- Founder-facing leadership. Translating business goals into architecture decisions, not just tickets.
| Role | Fractional CTO, Go Fractional |
| Location | New York, NY (Remote) |
| Since | June 2026 |
| Scope | Vetted Fractional CTO within the Go Fractional network, providing end-to-end technical leadership for venture-backed startups and high-growth founders |
- Own the entire product build from initial architecture to production-grade SaaS, aligning engineering with business outcomes, retention, and revenue growth
- Deep expertise across early-stage MVPs, full-scale platforms, GovTech, and ClimateTech
- Available for fractional, advisory, and interim CTO engagements
| Role | Company | Location | Duration |
|---|---|---|---|
| Fractional CTO | Go Fractional | New York, NY (Remote) | Jun 2026, Present |
| Chief Executive Officer | Enumsol LLC | United States | Jan 2024, Aug 2026 |
| Chief Technology Officer | DiveBridge LLC | Montana, USA | Sep 2020, Aug 2026 |
| Senior Software Engineer | Generative Genie | Montana, USA | Jan 2023, Dec 2024 |
| Full Stack Engineer | SUNT | Stockholm County, Sweden | Mar 2021, Dec 2022 |
| Junior Full Stack Developer | SUNT | Stockholm County, Sweden | Sep 2020, Feb 2021 |
| Full-Stack Development Intern | 100Devs | Los Angeles County, CA, USA | Jan 2019, Sep 2019 |
Role highlights
SUNT, Full Stack Engineer / Junior Full Stack Developer
- Led development of data-intensive, enterprise-grade web applications using React.js, Vue.js, Django, and PostgreSQL
- Built out core features of a renewable energy analytics platform handling millions of data points daily; integrated data visualization that reduced client reporting time by 50 percent
- Delivered 100 percent of sprints on time across Agile SCRUM cycles while maintaining high code quality
Generative Genie, Senior Software Engineer
- Led development of robust RESTful APIs, integrated with modern front-end technologies for optimal UX
- Managed and optimized databases and server infrastructure for high availability, scalability, and performance
- Architected and developed a secure authentication-backed application, with thorough codebase and API documentation
100Devs, Full-Stack Development Intern
- Completed an 8-month, project-based full-stack development program (React, Node.js, MongoDB) through daily coding exercises and pair programming
- Built and deployed responsive web applications from scratch, applying industry-standard version control, testing, and deployment practices; improved personal coding speed and efficiency by 30 percent
| Engagement | What I Did | Stack |
|---|---|---|
| Sparklyst | Hardened a Lovable-built ("vibe-coded") app for launch: tightened access control, fixed authentication and session handling, resolved root-cause state bugs | React, Lovable, Auth |
| SUNT — sunt.tech | Public-sector supplier rating platform for structured evaluation of 140,000+ suppliers across 30+ public organizations, cutting analytics processing time by 50 percent | Django, PostgreSQL |
| VentureMind — venturemind.ai | Full-stack SaaS integrating ML models for automated investment insights at 95 percent predictive accuracy | Next.js, Node.js, MongoDB, AWS |
| MIRRA | Secure, embeddable virtual try-on widget for Shopify, WordPress, and custom storefronts, from architecture through merchant-ready documentation | JavaScript, Shopify, WordPress |
Earlier portfolio work
| Project | Focus |
|---|---|
| Marketing.Biz | High-performance frontend, efficient state management, conversion-focused UI |
| DocuGuardian | Secure document management with advanced encryption and fast backend processing |
| Toksta | Intuitive dashboard with real-time updates and optimized navigation |
| JakMet | Metal trading platform with real-time pricing and automated inventory |
| OptiWork | AI-driven workforce management, predictive scheduling, 40 percent efficiency gain |
Full-Stack Engineering
Builds production React/Next.js and Vue/Nuxt frontends backed by Node/Express or Django services, with Tailwind and Framer Motion for polish, and Bubble.io or WordPress/Strapi when a no-code or CMS-driven layer is the right call. Designs REST and GraphQL APIs documented via Swagger/Redoc 8000 , modeled with Prisma, and manages state with Redux Toolkit across PostgreSQL, MySQL, MongoDB, Redis, and Supabase depending on the data shape.
System Design and Distributed Systems
Approaches architecture by separating data-intensive from compute-intensive workloads, translating requirements into a concrete scaling decision, and removing bottlenecks at the data-structure level before they reach production.
- API and protocol architecture: REST, GraphQL, gRPC, WebSockets, and SOAP, chosen to fit the job instead of defaulting to one
- Data layer design: relational versus NoSQL modeling by access pattern, hands-on with MongoDB, Cassandra, DynamoDB, and Redis
- Caching strategy: multi-layer caching with read-through, write-through, and write-back strategies, tuned eviction policy and TTL
- Load balancing: algorithm selection by traffic shape, active and passive health checks, JWT versus sticky-session trade-offs
- Replication and partitioning: single-leader and multi-leader replication, quorum-based consistency, key-range and hash partitioning
- Reliability and observability: CAP-theorem-grounded design, SLA/SLO targets, latency percentile tracking from P50 through P99.9
- Asynchronous messaging: queues and pub/sub with explicit ordering guarantees, idempotency, and dead-letter handling
- Applied case work: full video-streaming architecture, covering adaptive bitrate, transcoding, CDN placement, and capacity sizing
AI and Agentic Systems Engineering
- Context engineering: architects the full context window deliberately, system instructions, history, retrieved knowledge, tool schemas, and persistent state, rather than treating a prompt as one blob of text
- RAG pipelines: ingesting and chunking with fixed-size, sliding-window, semantic, and recursive strategies; indexing in FAISS, ChromaDB, Pinecone, or Supabase pgvector; dense plus sparse retrieval with cross-encoder re-ranking
- MCP and tool use: architects MCP clients and servers across stdio, SSE, HTTP, and WebSocket transports, with ReAct-style reasoning-and-acting loops
- Multi-agent systems: isolated sub-agents, fan-out/fan-in processing, and context-compression techniques for long-running workflows
- Agentic memory: tiered episodic, semantic, and procedural memory mapped to hot, warm, and cold storage layers
- Production infrastructure: persistent headless agent processes on cloud infrastructure, with autonomous messaging interfaces over the Telegram Bot API
- Reinforcement learning and voice AI: distributed GPU training with GRPO and process reward models; cascading and native audio-to-audio voice pipelines with dialect tuning
Claude Code and Agentic CLI Development
Treats AI-assisted coding as a discipline, not a shortcut, deliberately choosing between traditional coding, vibe coding, no-code tools, and spec-driven development based on what a task actually needs. Enforces a hard "done versus verified" bar: nothing ships without manual inspection of edge cases, UI states, database fields, and state transitions.
- Agentic workflow: operates agents inside a disciplined Think, Act, Observe loop; acts as senior architect while directing the agent as an execution-focused junior developer
- Environment and cost engineering: CLI setup across VS Code, JetBrains, and web shells, with active token-spend and billing management
- Permissions and security: directory trust boundaries, execution-mode tuning, and ignore rules that keep agents away from secrets
- Command-line fluency: full slash-command surface, from session control to automated planning, auditing, and state management
- Model strategy: matches model to task, Sonnet for daily coding, Opus for architecture-level reasoning, Haiku for fast scans, plus local open-weight models via Ollama
- Sub-agents and worktrees: YAML-defined sub-agents, parallelized work via git worktrees, and custom SKILL.md workflow automations
- Full-stack delivery: wiring frontends to live APIs, FastAPI backends with SQLAlchemy and Alembic, and server-side RBAC
- Git discipline: short-lived feature branches, end-to-end verification before merge, Conventional Commit formatting
Cloud, DevOps and Infrastructure
- Cloud: AWS (EC2, deployment infrastructure), Google Cloud Platform (Cloud Run, Cloud Build, Artifact Registry)
- Containers: Docker, multi-stage Dockerfiles, containerized services shipped to Cloud Run with live domains and secret bindings
- CI/CD: automated build and deploy pipelines, serverless deployments
- Version control: Git and GitHub, issue and PR lifecycles managed via CLI and MCP
| Credential | Issuer |
|---|---|
| B.S., Software Engineering, Sukkur IBA University (2018, 2022) | Sukkur IBA University |
| Google UX Design Professional Certificate | |
| Google AI Essentials | |
| Google IT Automation with Python Professional Certificate | |
| IBM Software Engineering Essentials | IBM |
| IBM Web Development Fundamentals | IBM |
| Introduction to IoT | CISCO |
Notes on owning technical builds end to end: architecture decisions, AI-to-production hardening, and what actually breaks after launch.
I'm best suited to founders who need full technical ownership, not just execution on an existing plan, especially in healthtech and proptech.
