A.I • FULL-STACK
About Me
Hi, I'm Tofara — a software engineer raised in Harare, Zimbabwe, now based between Dubai and Abu Dhabi. I really enjoy building software, and outside of work, I spend most of my time experimenting with AI and new technologies.
My job and career have required me to get great at AI, full-stack development, and infrastructure engineering — building production-grade systems that scale, adapt, and solve real-world problems. I've worked across the full stack, from designing APIs and scalable architectures to developing agentic AI workflows that automate complex reasoning tasks.
I'm all about the quality behind software — transforming ideas into reliable, intelligent systems. My work combines deep technical problem-solving with product mindset — performance, maintainability, and real-world impact.
What I Do
I've been engineering at fast-moving, high-tech startups, and I currently work as a Software Engineer at Mercanis, a German AI startup that raised $30M in Series A (2025).
My full AI engineering stack:
- Architecting agentic AI systems and event-driven microservices — designing multi-agent workflows with LangGraph and custom agent frameworks for autonomous reasoning, tool orchestration, and state management across distributed systems. Building production backend services with Node.js, NestJS, TypeScript, and FastAPI on AWS infrastructure.
- Integrating LLM APIs and building AI-native applications — working with OpenAI, Anthropic (Claude), Hugging Face Transformers, and Perplexity to build production systems with multi-model orchestration, streaming responses, function calling, and Model Context Protocol (MCP) integrations.
- Building full-stack web applications — developing real-time AI dashboards and interactive agent interfaces using TypeScript, Next.js, React, and Tailwind CSS. Implementing REST APIs with PostgreSQL, MongoDB, and Redis for data persistence and caching.
- Implementing infrastructure and observability at scale — deploying containerized services with Docker and Kubernetes, building CI/CD pipelines via GitHub Actions, and monitoring production LLM systems with event streaming and asynchronous processing.
At my core, I'm a Type 1 engineer — deep engineering fundamentals, able to learn what's needed to get the job done well. I care about every layer, from the database schema to the user experience, and the intelligence that connects them.
Ventures & Side Projects
Outside work, I'm building:
- Sylabl — hybrid search platform for Cambridge past papers using LLM embeddings + pgvector. 16,000+ papers indexed, 82 students on day one. Built to solve the broken-link mess I dealt with during my Cambridge exams.
- TLabs — AI-first software studio transforming how companies operate through intelligent automation and scalable system design.
Technical Expertise
I'm fluent in multiple layers of modern engineering:
- Languages: Python, TypeScript, C, C++, Go
- AI Dev Tools: Claude Code, Cursor, GitHub Copilot, Model Context Protocol (MCP), and prompt engineering
- LLM APIs & SDKs: OpenAI, Anthropic (Claude), Hugging Face Transformers, Perplexity, Groq, and Cohere
- Frameworks & Testing: FastAPI, Next.js, NestJS, React, LangGraph, CrewAI, pytest, Jest
- Infrastructure & Cloud: Docker, Kubernetes, AWS, Cloudflare Workers, Redis, RabbitMQ, PostgreSQL, MongoDB, GitHub Actions, Prometheus, Grafana
- AI Systems: RAG pipelines, vector databases (Pinecone, Qdrant), agentic workflows, LLM evaluation tools, and prompt engineering
- Automation: n8n, Pipedream
- Software Principles: SOLID, Domain-Driven Design (DDD), Test-Driven Development (TDD), and Clean Architecture
My foundation was built on two years of intensive C/C++ programming, where I developed HTTP servers, game engines, and multithreaded applications. This low-level grounding taught me how computers truly work — memory, processes, networking — and gave me the technical confidence to design complex, distributed systems.
I later transitioned to full-stack web and AI development, where I now blend modern web frameworks with machine intelligence, creating systems that learn, adapt, and scale.
Work & Collaboration
Beyond the technical layer, I bring strong communication and teamwork. I've worked in remote, multicultural teams spanning the US, Europe, UAE, and Africa — coordinating across time zones, aligning business priorities with technical execution, and delivering consistently.
I'm comfortable working directly with non-technical founders, PMs, and clients, translating high-level product ideas into concrete technical solutions. I also enjoy mentoring peers, reviewing code, and improving developer workflows through automation and DevOps practices.
My professional values are: clarity, reliability, and precision — whether in communication or code.
Education
- École 42 Abu Dhabi (Software Engineering Program)
Intensive peer-driven program focused on low-level programming, system design, and large-scale software projects. - Cambridge International A Levels: 1A*, 2A's, 1B (Mathematics and Physical Sciences)
- IGCSE: 8A*'s and 1A
- SAT Math: 770 / 800
These experiences trained me to think algorithmically, debug deeply, and approach problems with both creativity and rigor.
Vision & Craft
I believe the best software blends engineering precision with creative design. My long-term goal is to create systems that serve millions of users, combining robust infrastructure with human-centered AI capabilities.
I'm especially interested in how LLMs, agentic systems, and retrieval architectures can power the next generation of educational and productivity tools — systems that understand and adapt to human intent, not just execute commands.
Let's Talk
If you're building something ambitious — whether it's a startup, AI product, or open-source tool — I'd love to hear about it.
You can reach me via email or connect with me on LinkedIn.
Let's build something intelligent, useful, and lasting.