Portfolio & case studies · LinkedIn · Resume · Contact me
I am Akshit Pareta, an AI Native Product Manager based in Jaipur, India. My background combines commerce, consulting, and hands-on product delivery since 2019. I work with founders, customers, and operations teams to understand the business, choose what to build, and ship with AI tools.
At Gosotek Technology, my work includes a live FMCG customer and operations platform serving approximately 15,000 customer users and 100 staff. It connects B2B purchasing with fulfilment, accounts, and daily business operations.
I stay involved after launch: customer feedback, product changes, QA, release preparation, security, and production support. I am interested in AI Product Manager and Technical Product Manager roles with startups and growing businesses, plus contract, consulting, and project work.
| Product | Business problem and my work | Stage |
|---|---|---|
| FMCG commerce & operations | Connected customer purchasing with internal fulfilment and accounts across customer, operations, and web/admin apps. Used feedback to refine self-pickup incentives, quantity-based pricing, delivery choices, and payment-failure handling. Conversational admin actions use permission-scoped access and human confirmation before changes. | Live company project |
| Fashion stock clearance | Shaped a B2B marketplace around manufacturers' unsold stock and cash-flow needs, using direct supply, smaller purchases, volume buying, and pre-orders. Early business feedback is positive; outcome improvements have not been quantified. | Company project, early adoption |
| Minirak | Designed and built connected personal, office, and freelance workspaces with notes, tasks, team tagging, and progress tracking. | Personal product, pre-launch |
Explore the open-source apps and reusable AI workflows I publish below.
| Project | Problem it addresses | Status / license |
|---|---|---|
| Retail Buddy | A kirana POS and inventory app with GST billing, UPI, WhatsApp bill sharing, and offline-first workflows. | Public source · MIT |
| HiveDesk | An in-browser ERP sandbox exploring role-aware AI assistance and confirmation before writes. It is not a live Frappe deployment. | Sandbox · MIT |
| Kendra | An internal knowledge Q&A desk with retrieval, citations, and a bundled demo corpus; live-model answers are optional. | Public source · MIT |
| Vellum | An on-device PDF-to-Markdown tool for converting documents without uploading them. | Public source · MIT |
| Agent Harness System | A reusable orchestrator/worker harness for structured AI-assisted execution. | Public source · MIT |
| Product Positioning Skill | A reusable skill for product positioning, taglines, audience definition, and brand voice. | Public source · MIT |
Browse all public repositories · View authored public pull requests
- Understand: talk with customers and staff, map the business workflow, and identify where work gets stuck.
- Prioritize: define requirements, scope the first release, and agree on acceptance criteria.
- Build and ship: use AI-assisted development, integrations, testing, deployment, and release checks to deliver usable products.
- Improve: use feedback and operational issues to refine the product after launch.
My delivery work includes apps, automations, conversational assistants, API integrations, Cloudflare, VPS operations, and app-store releases. I use tools and systems to solve the business problem and stay responsible for the result.
Open to full-time roles, contracts, consulting, and project engagements.
Discuss a role or project · Visit my website · Connect on LinkedIn

