A decade embedding with strategic customers — from UX designer, to co-founder of a firm scaled to 22 product and engineering professionals, to shipping agentic AI into production. I've put a Claude-powered CMS live in a client's stack — where the ownership team now publishes market-ready content without waiting on developers — pair-coded in clinics and fintech pods, and carried field lessons straight back to product. Fin's Forward Deployed Software Engineer seat in Dublin is the same work under a clearer title: deep technical delivery, real customer impact, and influence on product direction.
“The best software I have shipped was never built from a clean spec — I lead the way to find the product direction with micro-demos, surfacing opportunities, and guiding the customer toward the path that works for their business.”
FFauzul Kabir ChowdhuryApplying for Forward Deployed Software Engineer · Dublin
Why Fin
Conviction over hype.
What pulled me toward Fin wasn't a feature list — it was the clarity of conviction running through the company, starting with Eoghan McCabe. I first came across Fin — and Eoghan himself — through his conversation on Lenny's Podcast, and that focus — raw, customer-obsessed, building actual utility instead of AI theatre — made me realise that's the room I want to be in. It maps directly to how I work: I don't close implementation gaps with slide decks; I lead the hacking phase of an ambiguous brief, drop into the customer's repo, and code side-by-side with their engineers until the system holds under production load.
Procedures & Tasks — code and live data, mid-conversation
Fin blends document-style steps with real code execution and live data connectors to handle refunds, cancellations and lookups inside the chat. That's the architectural tier I already build at — VisaPros coordinates parallel FastAPI country agents over Google's A2A protocol, and I run autonomous orchestration in n8n with custom MCP servers bridging LLMs to client systems.
The Fin AI Engine™ — retrieve, rerank, resolve
A multi-layer pipeline — RAG retrieval, a reranker, a summariser and an escalation classifier — is exactly the kind of system I reason about and harden: semantic chunking, eval suites before go-live, and routing by reasoning-depth versus latency rather than betting on a single model.
Built for regulated, high-stakes customers
Fin earns trust in fintech and e-gaming, where a wrong action is expensive. I've shipped instance-level OIDC/RBAC (Anygum) and HIPAA/GDPR pipelines across five organisations — so I can absorb the security and integration risk that surfaces the moment Fin touches a strategic customer's stack.
One agent across Service, Sales & Ecommerce
Fin's shift from a point-solution bot to a multi-role Customer Agent that changes role mid-conversation is the harder, more valuable problem. The generalist span I bring — design-led product sense, enterprise delivery, and agentic engineering — is built for exactly that breadth.
Scale & impact
A career optimized for shipping
22Product & engineering professionals at ELO
9Client geographies delivered
6AI-native products shipped solo (2025–26)
5Client orgs led into GDPR / HIPAA compliance
The Fin loop, the way I already build
Train · Test · Deploy · Analyze
Fin's flywheel is exactly the discipline I bring to every customer engagement — scope, prove, ship securely, then feed the field back into the product.
Fin · workflowSample
Knowledge sources
Scoping procedures & success criteria with the customer.
Discovery notes · customer workflows● Mapped
REST / webhook integration spec● Scoped
GDPR & HIPAA compliance map● Synced
3 sources ingestedsuccess criteria set
01
Train
Scope procedures, knowledge sources and success criteria with the customer — validate the ambiguous brief before a line of code.
02
Test
Simulate production conversations and run eval suites (Jest, Cypress, Playwright) to catch hallucination, latency and drift before go-live.
03
Deploy
Wire secure REST payloads, webhooks, OIDC auth and RBAC straight into the customer stack — auditable and reversible by design.
04
Analyze
Measure resolution, escalation quality and edge-case failures — carry field lessons straight back to Sales, Success and Product.
Point by point against the JD
14 reasons Fauzul fits this seat
Fin's Forward Deployed Software Engineers work at the intersection of deep technical delivery, customer impact, and product strategy — embedding with strategic accounts, coding in their stacks, and shaping what Fin ships next. Here is a point-by-point breakdown of how Fauzul's decade-long story maps onto the role.
01
Strong fit
Strong engineering background, preferably in fields such as Computer Science, Mathematics, Software Engineering or Physics.
Nine years of production software engineering across full-stack TypeScript/React, Go microservices, and Python AI pipelines — grounded in a UI/UX and product-design foundation that keeps implementation tied to human outcomes. Co-founded in 2017 and personally architected core frontend frameworks, design systems, edge SSR pipelines, and multi-tenant component architecture across the client portfolio before moving into CIO leadership.
02
Strong fit
Experience building a SaaS product with real-world usage. Bonus points for former founders or early startup engineers.
Fauzul co-founded in 2017 and scaled it from a blank page into a 22-person product product engineering venture firm. In this seat, he owned the delivery of enterprise-grade SaaS products used daily by global end-users — including , where he created the first web product and admin panel for AI-powered loop matching from a blank canvas, stood up the full-stack team on an NX monorepo after initial market success, and saw the product grow over six years into web, DAW plugin, and desktop surfaces before Loopcloud acquired it in 2026 (the dedicated team still ships modern integrations). Other portfolio proof includes health-tech ( — 13,340 clinical schedules, 3,242 patients onboarded in 2026) and enterprise marketplaces (). Over the past year he has also operated as a high-velocity solo builder, shipping six AI-native products (, , , CMS, , and ). He brings the raw, self-directed ownership, business judgment, and operational grit that only a startup founder possesses.
03
Strong fit
You're interested in AI/ML and enjoy quickly building a strong understanding of novel problems, capabilities, and architectures — including the trade-offs and limitations involved.
He is deeply embedded in the AI/ML frontier, actively building and experimenting with multi-agent systems and custom harnesses. In his Google × Kaggle AI Agents intensive capstone (2026), he shipped —a multi-agent visa advisor utilizing Google's advanced ADK Agent-to-Agent (A2A) hub-and-spoke protocol. He is highly active in the developer ecosystem, evaluating the trade-offs of emerging standards like AP2 (Agent Protocol 2) and Model Context Protocol (MCP) for bridging data stores to LLMs. He understands LLM constraints (prompt caching boundaries, latency-cost curves, context window degradation, and reasoning limits) and has engineered mitigation frameworks like semantic chunking hybrid models and asynchronous evaluation guardrails to ensure reliability in production.
04
Strong fit
You bring strong cross-functional collaboration skills, with the ability to navigate organizational boundaries and bring people together to achieve the best outcomes for our customers and for Fin.
Starting his career in UI/UX and product design gave Fauzul a natural cross-functional vocabulary. He does not see walls between engineering, design, and GTM; he sees a single product experience. As CIO at , he sat in the middle of engineering squads, product managers, and enterprise buyers for nine years. He excels at translating complex technical limits into strategic business outcomes, aligning client-side security teams, procurement officers, and GTM stakeholders. He has successfully guided five enterprise organizations into GDPR, HIPAA, and WCAG 2 accessibility compliance, proving his ability to navigate high-stakes organizational boundaries to secure successful deployments.
05
Strong fit
You're a strong coder. You have deep knowledge of at least one high-level programming language (for example, JavaScript, Python, Ruby, etc.). It doesn't need to be a language we use today — great people are effective learners and are quick to pick up new tools or introduce better ways of working.
Fauzul possesses deep production-grade knowledge of TypeScript/JavaScript, Go, and Python. He utilizes TypeScript for building responsive, type-safe React/Next.js interfaces and robust Node.js backends. He writes Go for highly concurrent, low-latency microservices, and utilizes Python for his AI agent work, data pipelines, and custom n8n orchestration nodes. He approaches programming with a language-agnostic discipline, focusing on typed contracts, strict CI/CD gates, automated unit/integration test suites (Jest, Cypress, Playwright), and clean, self-documenting code.
06
Strong fit
You're able to prototype and ship software quickly, while maintaining a high bar for quality and a clear focus on business goals and outcomes.
His UX design foundation is a major accelerator for his prototyping speed. He can wire up highly polished, interactive React interfaces and prototypes in a matter of hours, allowing stakeholders to touch and validate an experience before a heavy backend is even built. He combines this visual speed with engineering rigor: at , he drove a 4x faster time-to-market (75% reduction in cycle time) using custom CI/CD pipelines while keeping security and compliance baked into the architecture. His builds are laser-focused on business outcomes: for , he took a Claude-powered CMS from a loose natural-language brief to a production-live edge deployment in weeks — and the client team now moves from thought to market-ready content without waiting on developers or routing brand changes through a business desk. The legacy-to-AI shift became a genuine eye-opener for how they operate day to day.
07
Strong fit
You're proficient in using and writing REST APIs to integrate systems effectively.
Deeply familiar with diverse integration protocols. He has designed and delivered API systems across REST, GraphQL, backend microservices with gRPC, and real-time WebSockets, managing complex authentication like 's federated OIDC/RBAC gateways. He has implemented secure transaction idempotency for payment services (Stripe at ) and coordinated Norwegian payment integrations (Vipps at Frontgo), writing custom MCP servers to allow LLMs to safely query and execute actions on client-side databases over secure endpoints.
08
Strong fit
You're a strong communicator who can comfortably discuss the same topics with engineers and non-technical audiences, adapting depth and detail as needed.
Translating ambiguous customer needs into concrete engineering plans is his default operating mode. He easily switches altitudes, discussing ROI, security risk, and deployment timelines with C-suite executives, and API schemas, websocket states, and webhooks with customer engineers. Beyond core engineering, he drives creative social media content creation—running the 'French_with_Fauzul' learning channel and fashion-focused content. This public writing, developer documentation, and creative media production—combined with his consistent, self-directed 900+ day Duolingo French streak—demonstrates his structural discipline, love for clarity, and long-term communication consistency.
09
Strong fit
You're willing to travel and work on-site with customers to build strong relationships and develop a deep understanding of their needs.
He is highly intentional about basing this career chapter in Dublin — Fin's EMEA hub, Critical Skills Employment Permit route ready, with family in the UK and close ties in Norway. A decade of remote-led enterprise delivery across London, Zurich, Oslo, Dubai, Florida and Montreal — multi-year partnerships sustained through disciplined discovery and reliable delivery, with on-site embedding where the work demanded it (weeks in clinics; in-person client meetings in Montreal). He is fully open to — and energized by — customer travel and on-site embedding as a Forward Deployed Engineer. He has publicly championed this exact customer-centric, travel-oriented FDE mindset (such as his viral post discussing Fin's own FDE organizational philosophy, which resonated directly with Des Traynor, Fin's official brand account, and Fin's product leadership team). He knows that real adoption is won and actual customer pain is observed firsthand.
10
Adaptable
Drive the adoption of Fin by helping prospects unlock the power of Fin's AI capabilities to automate and scale their support operations.
While he has successfully driven AI adoption with legacy enterprise clients — at , replacing a traditional CMS with a Claude natural-language engine the ownership team now uses to publish market-ready content without developer queues — and has sold complex technical solutions for a decade, he has not yet deployed Fin's specific platform or worked inside a dedicated customer-service/helpdesk GTM pipeline.
How I close it
He approaches adoption by making 'AI tangible' for prospects. By building interactive, high-fidelity prototypes and real-time evaluation dashboards (similar to our stateful mock widget), he breaks through stakeholder skepticism. He will leverage Fin's Train → Test → Deploy → Analyze loop from week one—building trust by showing prospects actual simulation evals and precise resolution-rate telemetry before going live. He will close the helpdesk domain terminology ramp (SLAs, CSAT, escalation parameters) within 2 to 4 weeks.
11
Strong fit
Embed deeply with strategic customers to understand their business challenges and technical requirements, and provide the insight needed to achieve outstanding results with Fin.
Understanding the real business challenge—not just the written brief—is his default. For 's clinical rollout he didn't sit in a conference room; he spent weeks on-site in specialized diabetes and wound-care clinics, observing doctors and nurses use the software under pressure to catch friction the requirements doc never captured, then turned those field observations into concrete product direction. As CIO at he did this across health-tech, fintech, and enterprise marketplaces for nine years — learning each customer's business deeply through discovery and delivery across geographies, embedding on-site when the work demanded it, then advising on what would actually move their outcomes.
12
Strong fit
Work hands-on with customers' technical teams as a trusted technical expert, coding side-by-side to drive projects through to completion.
This is his default operating style, not an aspiration. For Norway's Frontgo, he embedded directly with their local engineering pod to pair-code a secure Vipps fintech integration and establish modern, auditable testing practices in their codebase, acting as a trusted technical advisor from planning to final release. He codes side-by-side with client teams rather than handing over a spec and walking away—it is the fastest, most reliable way to drive a complex integration all the way to completion.
13
Strong fit
Collaborate closely with Sales, Success, and Product teams to ensure seamless customer experiences, successful project delivery, and actionable product feedback.
As CIO and co-founder, Fauzul was the product feedback loop. He knows how to synthesis messy field lessons into structured, actionable product telemetry for engineering teams (such as translating client-side auth friction into 's standardized OIDC backbone). He knows how to give Sales exactly what they need to close enterprise deals: highly polished, custom-branded POCs, clear mitigation paths for security/compliance audits, and precise ROI metrics.
14
Strong fit
Contribute to the evolution of the Forward Deployed Software Engineering function by codifying best practices and sharing insights from customer engagements to help the team scale and grow.
Fauzul built 's entire product delivery and engineering playbook from scratch as they scaled to 22 engineers (52+ professionals total FT & contract). He codified their API integration standards, security and HIPAA compliance checklists, and engineer onboarding rituals. Furthermore, he is a dedicated mentor—he has coached engineers into their best next roles, and alumni now lead at organisations in Norway (2), Canada, Germany, plus two leading local Bangladeshi tech companies. He has the exact organizational muscle needed to codify Fin's founding FDE playbooks and scale the team's footprint.
Ask Fauzul's AI
Capabilities assessment
Strengths & honest gaps
A balanced look at how the background maps to Fin's bar — and exactly how the gaps close.
A generalist story: design-led, business-honed, AI-driven
He started in UI/UX and product design, co-founded and scaled a 22-person product engineering venture firm, and evolved into a pioneer of agentic AI. This generalist journey gives him a rare combination of visual polish, enterprise business acumen, and raw engineering depth.
Agentic AI in production & frontier protocols
He doesn't just build chatbot demos; he ships agentic AI that enterprises trust and actually adopt. runs a Claude-powered CMS in production — the ownership team rejoiced once they could go from thought to market-ready content without waiting on developers — alongside emerging protocols like Google's A2A (Agent-to-Agent), AP2 (Agent Protocol 2), and custom MCP servers.
Deep integration, identity, & compliance scar tissue
He has guided 5 enterprise organizations through high-stakes HIPAA, GDPR, and security audits. He speaks federated OIDC, instance-level RBAC, secure webhooks, and REST/GraphQL payload optimization fluently, ensuring AI integrations are secure and auditable.
Rapid high-fidelity prototyping with a product lens
Thanks to his design background, he designs and codes highly polished, interactive frontends and proof-of-concepts in hours. This allows strategic customers to experience and trust an AI solution visually before committing heavy engineering resources.
Closing the gaps
2–4 weeks
No direct Fin / Intercom platform or customer-service SaaS deployment experience.
The FDE discipline is platform-agnostic: discover, scope, integrate, test, and deploy. He will ramp on Fin's specific Procedures, knowledge pipelines, and helpdesk schemas by running a simulated pilot engagement in week one, collaborating with Fin's AI Group, and translating his deep integration muscle directly to the customer-support domain.
Immediate
Production AI has been Claude/Gemini-led — not Fin's proprietary models.
Fauzul is honest and intentional: his shipped agents are Claude-led, with Gemini, inside complex harnesses — not Fin Apex. Because the FDE seat is focused on custom integrations, evaluation harnesses, and customer business outcomes—not training base models—his multi-model engineering and prompt-routing experience is a direct, immediate asset.
Ask Fauzul's AI
In production today
Fin · workflowSample
Knowledge sources
Scoping procedures & success criteria with the customer.
Discovery notes · customer workflows● Mapped
REST / webhook integration spec● Scoped
GDPR & HIPAA compliance map● Synced
3 sources ingestedsuccess criteria set
Agentic AI, already earning customer trust
Forward Deployed Software Engineering is where I have been heading — embedded with strategic customers, coding in their stack, and shipping AI that non-technical owners actually adopt.
A Claude-powered CMS, running in a client's production stack
For — a 26-year-old Dhaka-based export house serving global importers — I skipped the traditional headless CMS. The client edits the live site in plain English; Claude opens a PR; CI/CD ships it to production on Cloudflare Workers. Natural-language intent in, reviewed and reversible change out — and the ownership team can now go from thought to market-ready content without waiting on developers or routing every brand change through a business desk. For a legacy export house, that self-serve independence has been a genuine eye-opener.
Plain-English intent → Claude-authored PR → automated CI/CD deploy on Cloudflare Workers — real agentic AI in production today.
Every change stays inside Git — reviewable, auditable, revertible. No black-box edits to a live customer-facing surface.
Client adoption proof: the team rejoiced once they could refresh brand offerings on the spot — no developer queue, no business-team handoff.
Agentic 0→1 in public: , (Milan AI Week 2026 — results pending), — multi-step agents with A2A and MCP.
Compliance as product surface — HIPAA (), GDPR/VAT (), OIDC/RBAC () — before AI touches sensitive data.
Train — scope & success criteria
Ten years validating ambiguous briefs before over-building — pop-up demand tests for Haiba, granular tickets from executive conversations.
Test — eval before go-live
Technical evals, POCs, and automated test suites (Jest, Cypress, Playwright) — the discipline to simulate before customer-facing deploys.
Deploy — REST & identity
OIDC/RBAC at Anygum, Vipps at Frontgo, Stripe at Jamahook — integrations that keep customer data isolated and auditable.
Analyze — field → product
Nine years carrying client lessons back into platform decisions — the feedback loop Fin's FDE needs between R&D and GTM.
Working with models & agents
How I deploy agentic AI in production
I am model-agnostic by design. The FDE seat is about custom integrations, evaluation harnesses, and customer outcomes — not training base models — so my multi-model, multi-harness experience maps directly onto Fin's stack.
Multi-model & harnesses
Frontier models across diverse harnesses
I actively build and ship with Claude (lead) and Gemini inside different harnesses (Cursor, Bolt, Firebase firm), choosing the right model per task and routing by reasoning depth vs. latency rather than betting on a single vendor.
Agent protocols
A2A, MCP, and Google ADK in practice
For VisaPros I used Google ADK's agent-to-agent (A2A) hub-and-spoke architecture to coordinate six parallel country agents as FastAPI microservices. I write custom MCP servers to let LLMs safely query and act on client systems, and I actively track AP2 (Agent Protocol 2) as agent coordination standardizes.
In production
Agentic AI shipped, not demoed
A Claude-powered natural-language CMS runs in 's production stack; NewScriber productizes n8n-orchestrated autonomous briefing distribution; TagRamp embeds LLMs into a live workflow — real agentic systems with humans in the loop.
Evals & guardrails
Discipline before go-live
I scope success criteria first, then prove reliability with simulated conversations, automated test suites (Jest, Cypress, Playwright), and pre-token compression / prompt-caching strategies to keep latency and cost honest — the same Train → Test → Deploy → Analyze loop Fin runs.
Candid dialog
Answering the hard questions
Candid answers on why a co-founder and CIO wants a customer-embedded engineering seat at Fin, how a design and compliance background maps onto enterprise support automation, and a realistic plan for ramping on Fin's proprietary stack in Dublin.
You co-founded and ran your own product engineering venture firm — why a Forward Deployed Software Engineer seat at Fin?
Because customer-embedded engineering is the work I love most, stripped of agency management overhead. As grew, my daily role drifted toward running a client-services business. The part that lights me up is sitting with a strategic customer, understanding an ambiguous automation problem, and writing the code that solves it — then feeding what I learned back into product. Fin lets me do that with a frontier AI engine behind me, alongside R&D and GTM teams who want field signal, not folklore.
You haven't worked in customer service or helpdesk SaaS. Why should we trust you with Fin's strategic customers?
Because FDE is a deployment and custom-integration engineering role, not a tier-1 support seat. The job is embedding with a customer's technical team, scoping high-value workflows, designing secure REST/GraphQL payloads, and coding integrations until Fin drives measurable resolution. I have done exactly this in health-tech ('s clinical EHR), fintech (Frontgo's Vipps integration), and global export ('s AI CMS). Helpdesk terminology—SLAs, CSAT, ticket deflection parameters—is a vocabulary ramp I will close in two weeks; building trust with enterprise engineering teams and unblocking complex API contracts took me a decade to master.
Your production AI uses Claude, not Fin. How do you ramp on our stack?
I will never pretend to have shipped Fin Apex—I have not. But I learn AI architectures and developer tools incredibly fast. I went from zero to production builds with Google's ADK Agent-to-Agent (A2A) protocol, Model Context Protocol (MCP), and multi-model prompt routing within weeks. Fin's Train → Test → Deploy → Analyze cycle is exactly how I build: establishing clear eval harnesses before go-live, measuring token latency, scoring resolution rates, and carrying structured feedback back to Product. I will pair with Fin's AI Group from week one and build trust with clients by being transparent about my ramp, letting them see the engineering rigor firsthand.
Your career started in UI/UX design. How does that help you as a Forward Deployed Software Engineer?
It is my unfair advantage. Starting in UX built an obsession with detail, polish, and customer empathy. In the field it means I don't just write API pipelines — I design how automation is surfaced, making complex agent behaviour approachable to non-technical client teams. I can wire polished interactive POCs in hours so stakeholders touch the experience before heavy engineering commits. Empathy for the end-user is how AI actually gets adopted.
What is your experience with frontier agent protocols like Google's A2A or AP2?
I am deeply fascinated by how agent coordination is standardizing. In my build, I used Google's Agent-to-Agent (A2A) hub-and-spoke architecture to coordinate a central 'Advisor' agent with specialized 'spokes' that handled document validation and criteria matching. I am actively tracking AP2 (Agent Protocol 2) and MCP because I believe standardization is the key to scaling enterprise AI. At Fin, having an FDE who understands how to coordinate multi-agent teams and write custom MCP servers is a massive asset as clients ask to connect Fin to highly customized internal tools.
How do you handle loose, highly ambiguous briefs from enterprise clients?
By writing the discovery playbook. Enterprise clients often have a symptom ('our support queue is backed up') but don't know the cure. I handle this by embedding with their technical and support teams, mapping their actual workflows, and creating high-fidelity interactive prototypes. I translate their ambiguous intent into scoped, measurable success criteria (e.g., 'deflecting 40% of billing queries with zero latency'). Prototyping and aligning expectations visually and contractually is the only way to avoid scope creep and ensure final delivery success.
Why Dublin — and are you set up for hybrid in-office work?
Dublin is the natural EMEA hub for this work — Intercom/Fin already runs a mature engineering org here with CSEP sponsorship infrastructure, and the timezone overlap with my existing EU client delivery is zero-cost. I'm intentional about relocating for hybrid three-days-in-office and customer travel from day one. On a personal level, my sister lives in the UK and my best friend is in Norway — Dublin puts me closer to both. I have delivered remotely for clients across London, Oslo, Zurich, Dubai, Montreal, and Florida for a decade — with on-site work at and in-person meetings in Montreal — and I'm ready to embed on-site with Fin's strategic customers from Dublin.
Ask Fauzul's AI
Beyond the Role
What else Fauzul brings to the room
Capabilities and leadership scar tissue that extend far past the core job description.
UX & Design Engineering Bridge
Taste
Rare dual-competency from years as a UX Engineer. He bridges Figma layouts to React codebases with pixel precision, creates custom SVG assets, and designs micro-animations that make demos and products feel premium.
Developer Mentoring & Playbooks
People
As 's CIO, he established the studio's engineering standards and playbooks. He has guided junior developers into staff roles — with alumni now leading technical teams in Norway, Canada, and Germany.
Compliance & Audits by Design
Trust
Brings hands-on experience guiding five organizations through GDPR, HIPAA, and WCAG 2 audits. He treats privacy and access control (OIDC/RBAC at ) as a core product feature, not a checklist.
Founder & CIO Commercial Grit
Strategy
Eight years of co-founder operator experience. He understands run rates, value-based scoping, and resource prioritization. He partners with product and sales to drive revenue, not just deploy code.
Track record
The stack & the builds
Production-grade builds and integration patterns that prove Fauzul can handle complex API contracts, agentic orchestration, and enterprise security boundaries from day one in the field.
Forward Deployed & Customer-EmbeddedStrategic customer embedding · Coding side-by-side with client teams · Discovery → pilot → production · On-site delivery & travel · Cross-functional GTM (Sales · Success · Product) · FDE playbook codification
AIAgentic Development · Agentic Workflow · LLM Integration · MCP · A2A · Google ADK · Prompt Engineering · Eval / harness hill-climbing · Token & cost-aware context engineering · Firecrawl · Cursor · Bolt · Firebase Studio · Nano Banana · Claude Code · Claude Skills · Claude Code Subagents · Claude hooks & tools · Anthropic Claude · Google Antigravity (agy CLI) · Gemini Spark (email topic loops) · On-device Gemma 4 E2B (this site chat · WebGPU + MediaPipe) · Kimi via OpenRouter (NewScriber) · Gemini TTS via Azure (NewScriber)
Editorial export site for Global Jute Trading Ltd — a Dhaka-based sourcing & exporting house with 26 years as a company and 35 years of jute trading under its principals. Built to support their push into deeper international markets, with a particular focus on Canada.
Strategic health-tech partnership and equity investment in Bangladesh's most advanced clinical practice for wound care, diabetes, and nephrology. Spearheaded high-utility, HIPAA-compliant EHR clinical dashboards and scaled digital patient advocacy.
Anygum is a working marketplace-style platform where anyone can host their own API and builders can compose bespoke apps that pull services from multiple vendors. Identity runs through Cloudsight — custom auth on Auth0 — so provider and consumer context share single sign-on, RBAC and org controls, multi-instance of the same app for different teams, easy telemetry, and a single payment funnel for consumption. Ideation started as an internal auth problem for SaaS distribution; it was spun out as Anygum. Elobooks.net uses it. Distribution limited wider market reach; the product itself works.
A new LinkedIn for orgs that actually deal. TagRamp connects organizations from goals to meaningful business deals through AI-mediated intros. Built solo during the World's Largest Hackathon (May 2025) by Bolt.new and submitted to Y Combinator; now scaling with active enterprise pilots and CXO-level partnership discussions.
Agentic news scraper & editorial audio briefing network. Scrapes tech and business news via Firecrawl, curates & scripts dual-host dialogues with Kimi via OpenRouter, and renders high-fidelity multilingual voices via Gemini TTS on Azure.
Submitted as the capstone for the Google × Kaggle 5-Day AI Agents Intensive (2026). VisaPros is an agentic visa eligibility advisor utilizing a Google ADK agent-to-agent (A2A) hub-and-spoke architecture — parsing documents once and running six parallel country agents as FastAPI microservices to evaluate fit across destinations.
Ask Fauzul's AI
Next steps
Bring Fauzul into the room — in Dublin.
Intentional about Dublin, Ireland — Fin/Intercom's EMEA engineering hub, Critical Skills Employment Permit route. Remote and hybrid client delivery across nine geographies; Dublin hybrid and customer travel are exactly what this Forward Deployed seat calls for from day one.
On a personal note: my sister lives in the UK and my best friend is in Norway — I'd like to be closer to both, and Dublin is the right base for that while building this chapter in EMEA.
