Jona Joy
I build AI systems, distributed backend platforms, and customer-facing products from whiteboard to production - with founding-engineer ownership.
Founding Engineer · AI & Agentic Systems · Backend & Distributed Systems · AWS · FinTech
3+
Years Building
70%
Faster Delivery
99.5%+
Core Uptime
0→1
Founding Engineer
01 / ABOUT
Founding engineer.
Still building.
Jona Joy
Kochi, Kerala, India
I'm a founding-minded Software Engineer building at the intersection of AI systems, distributed backend infrastructure, fintech platforms, and customer-facing products. I build from zero to production — architecture, APIs, cloud infrastructure, product flows, reliability, and the parts that have to survive real users and business pressure.
As one of the first three engineers at Kireap, I've spent 3+ years in early-stage ownership: product decisions, backend architecture, AWS infrastructure, customer conversations, production incidents, pricing systems, payment/PSP integrations, blockchain traceability, and shipping. I've built across Python/FastAPI, PostgreSQL, Redis, AWS Lambda, SQS, SNS, EventBridge, API Gateway, Cognito, LangGraph, RAG, OpenAI/Claude APIs, Hyperledger Fabric, and full-stack applications.
My strongest lane is where AI meets serious engineering: agentic systems, conversational AI, backend platforms, fintech, cloud-native services, and high-ownership product execution. I care less about demos and more about systems that solve messy operational problems in production.
PROFILE
EDUCATION
GPA 9.33 / 10B.Tech, Information Technology
School of Engineering, CUSAT
Class of 2024
WHAT DRIVES ME
- →AI systems that solve messy real-world problems, not just polished demos
- →Founder-level ownership: product, users, architecture, shipping, and outcomes
- →Distributed backend platforms that hold under production pressure
- →Fintech, payments, cloud infrastructure, and agentic AI systems with real business value
- →Learning deeply across ML/DL, model inference, and production AI engineering
FOUNDER STORY
0 → 100Engineer #2 · Kireap · 2023 – present
In 2023, as a 3rd-year engineering student, I joined Kireap as one of the first three engineers with one goal: build real products from zero. The role quickly became more than writing code — I was part of product thinking, architecture decisions, field demos, customer discovery, supplier discussions, dry runs, production launches, and business trade-offs. We worked like a founding team: small group, high trust, direct user feedback, and fast iteration. Between field work, customer conversations, MVP launches, backend systems, payments, traceability, AI/ML, and cloud infrastructure, I learned to own outcomes instead of tasks. That journey shaped how I build today: understand the user, make the right trade-offs, ship fast, and keep the system working under real conditions.
Full founder journey on LinkedIn →02 / FEATURED WORK
Projects that
ship and scale.
SprintPilot AI
Agentic project intelligence for engineering teams
Agentic project intelligence platform that reasons across engineering context — tickets, docs, project updates, and status signals — to generate useful project answers and reports. Built as a production-style AI backend with RAG, tool-calling, MCP integrations, evaluation workflows, and observability tracing.
- ·Multi-agent workflow with LangGraph, tool-calling, and project-context retrieval
- ·RAG pipeline over engineering knowledge sources with evaluation and traceability
- ·Observability-first backend with traces for agent reasoning, retrieval, and tool usage
MedVoice Agent
Conversational voice AI for healthcare workflows
Multilingual voice agent project focused on healthcare-style workflows where users can speak naturally to book appointments, check availability, or interact with structured backend data. Still in development, but designed around the hard parts of production Voice AI: speech quality, latency, hallucination control, conversation state, model inference flow, and reliable backend execution.
- ·End-to-end voice pipeline: Whisper STT, LLM reasoning, TTS, async FastAPI backend, and React frontend
- ·Structured intent routing for appointments, availability, and patient-style workflow interactions
- ·Guardrails to avoid hallucinated records by grounding responses in verified backend data
Code-Switch ASR Benchmark
Malayalam-English ASR benchmark for real multilingual speech
Built a Malayalam-English code-switching ASR benchmark to evaluate how speech models handle real users who naturally mix languages while speaking. The project focuses on ASR evaluation, strict WER/CER measurement, language-span metadata, switch-point analysis, and improving transcription quality for multilingual conversational AI systems.
- ·Created a 2,800+ clip human-reviewed Manglish speech benchmark with transcripts, language spans, and switch-point annotations
- ·Benchmarked Sarvam Saaras and Whisper models using strict WER/CER metrics for code-switched speech
- ·Published dataset and fine-tuned Whisper-small Manglish model checkpoint for ongoing transcription improvement experiments
ClariFi
Personal finance intelligence with explainable AI
Personal finance and credit intelligence platform combining deterministic financial calculations with AI-assisted explanations. Built to show how AI can support real financial decision workflows without inventing numbers, hiding the math, or breaking user trust.
- ·Budgeting, forecasting, credit scoring, and AI-assisted financial insights in one workflow
- ·Deterministic calculation layer keeps financial outputs explainable and auditable
- ·FastAPI + PostgreSQL backend with React frontend and production-style workflow design
Satellite Change Detection
Computer vision pipeline for bi-temporal satellite imagery
Built for the DPIIT Hackathon organized by IIT Delhi — an ML pipeline for detecting human-made changes in satellite imagery using time-separated image pairs. This project demonstrates model inference, computer vision architecture, backend serving, and applied AI beyond LLM workflows.
- ·Siamese Transformer architecture for comparing bi-temporal satellite image pairs
- ·FastAPI inference service with preprocessing and evaluation/demo workflow support
- ·Designed to detect construction, roads, demolition, and human-made change patterns
PRISM Finance
On-chain lending with reputation-based credit logic
Decentralized lending prototype exploring how reputation scoring and DAO-backed collateral pools can reduce over-collateralization in SME-style credit flows. Built as a full product architecture across smart contracts, frontend, and transaction flow.
- ·Designed smart-contract architecture for reputation-based lending logic
- ·Implemented DAO-backed collateral and automated loan issuance flows
- ·Connected Solidity contracts to a production-style Next.js frontend
InvoChain
Invoice tokenization for SME liquidity
Web3 invoice factoring platform enabling small businesses to tokenize receivables and access liquidity through investor marketplaces. Included smart contracts, backend services, and transaction workflow design.
- ·ERC-compatible invoice token contracts with settlement logic
- ·FastAPI + PostgreSQL backend for transaction and user management
- ·Secure buy/sell flows with payment distribution logic
FitMate
Pose tracking and calorie estimation with real-time ML
Fitness tracking app using ML Kit pose detection to track body movement in real time, count reps, estimate calories burned, and log workouts without manual input.
- ·ML Kit pose estimation for real-time body landmark tracking and rep counting
- ·Calorie estimation engine using pose data, movement velocity, and exercise classification
- ·Flutter frontend with FastAPI backend and PostgreSQL for session history and progress tracking
03 / EXPERIENCE
Where I've
built things.
Software Engineer — Core Technical Team
Kireap Technologies
- →Built and deployed an agentic software delivery platform using Claude, OpenAI APIs, LangGraph, and RAG workflows that automated specification generation, implementation planning, code scaffolding, testing, debugging, and deployment; enabled engineering teams to deliver full-stack applications in minutes and reduced feature delivery cycle time by ~70%.
- →Designed and deployed production AI orchestration workflows incorporating tool-calling, retrieval pipelines, structured outputs, observability, and evaluation frameworks to improve reliability, traceability, and operational readiness.
- →Led the pricing platform end-to-end, partnering directly with the founder on product, algorithmic, and cost optimization decisions; designed pricing microservices, dynamic pricing models, and automated SageMaker retraining pipelines, increasing user price acceptance by ~50% and improving pricing outcomes by ~65%.
- →Owned payment and PSP integrations end-to-end, including payout workflows, webhook processing, reconciliation, fraud prevention, and operational monitoring for a fintech platform processing EUR 100k+/month; developed an ML/NLP-based name-matching system for beneficiary verification and risk reduction while maintaining zero critical payment incidents across 14 months.
- →Built and maintained AWS cloud infrastructure using Lambda, SQS, SNS, EventBridge, API Gateway, Cognito, Terraform, and GitHub Actions, maintaining >99.5% uptime across core services.
- →Implemented authentication, authorization, OAuth/JWT flows, middleware, Cognito-backed identity flows, API Gateway integrations, and secure service boundaries across production backend systems.
- →Led a monolith-to-microservices migration using clean architecture and dependency injection, improving system throughput by ~2× and enabling independent deployments across engineering domains.
- →Migrated a distributed traceability platform from a public blockchain network to a permissioned Hyperledger Fabric deployment using Go, Node.js, and AWS Managed Blockchain; built monitoring and observability tooling for transaction validation and system health, improving throughput by ~50% while enforcing strict data privacy requirements.
- →Built observability and monitoring systems using structured logging, distributed tracing, dashboards, and alerting, reducing debugging and incident resolution effort by ~60%.
- →Shipped backend services and product features across the full SDLC, collaborating with founders, stakeholders, and end users to define requirements, evaluate technical trade-offs, design APIs and PostgreSQL schemas, deploy to production, and iterate based on customer feedback.
Software Engineer Intern — Founding Engineer
Kireap Technologies
- →Joined as engineer #2 of 3 at a globally distributed startup, contributing across product development, system design, customer pilots, and production deployments from MVP through scale-up.
- →Worked directly with founders, customers, suppliers, and field users through discovery sessions, dry runs, field demos, and 0→1 launches; translated ambiguous real-world feedback into technical solutions and product iterations.
- →Participated in on-ground dry runs and user discovery, including field conversations in Madhya Pradesh with farmers and stakeholders, helping connect product decisions to real operational constraints.
- →Built full-stack applications from scratch using FastAPI, PostgreSQL, AWS, Next.js, and Flutter, delivering APIs, customer-facing web and mobile applications, and administrative platforms that formed the foundation of the company's initial production systems.
- →Shipped 5 production applications across supplier, logistics, customer, traceability, and administrative workflows, owning delivery end-to-end from requirements gathering through deployment and operational support.
- →Established engineering standards including CI/CD pipelines, testing practices, code quality tooling, documentation standards, operational runbooks, and development workflows, improving engineering velocity and reducing review cycles by ~35%.
04 / SKILLS
Built with the
right tools.
Languages
AI & Agent Systems
Conversational AI & Speech
Backend & Infra
Cloud & DevOps
FinTech & Product Systems
Blockchain
ML / DL
Product & Startup
Frontend & Mobile
Coursework / Certifications
Always learning, always shipping.
Currently deepening: agentic AI systems, conversational AI, ASR evaluation, distributed backend performance, and production-grade AI platform engineering.
05 / CONTACT
Let's build
something real.
Whether you're hiring for founding engineering, AI backend systems, distributed platforms, fintech, or conversational AI — I'm interested in teams that ship fast, trust ownership, and build around real user problems.
Open to opportunities
Particularly interested in founding engineer, AI backend, AI platform, conversational AI, distributed systems, and fintech roles. I work best with teams that move fast, trust ownership, and care about production outcomes.
Send a message →