Headline

Backend Software Engineer @ Open Climate Fix | Python, Django, FastAPI, Go | AWS, Kubernetes, Microservices | PostgreSQL, RabbitMQ, Kafka, Redis | Ex-Orange Health Labs, Tata Communications (formerly Kaleyra)

About

Backend software engineer with 4 years building high-throughput Python services in production. I'm currently a Backend Software Engineer at Open Climate Fix, working on the backend systems behind solar generation forecasting. Before this I was at Orange Health Labs and Tata Communications (formerly Kaleyra), where I built FastAPI services handling thousands of requests per second, cut chatbot latency by 33%, and migrated order search from RDS to Elasticsearch for a 96% improvement in query performance. Most of my work sits at the intersection of API design, distributed systems and performance. Day to day that means Python (Django, FastAPI, Flask) and Go, backed by PostgreSQL, MongoDB, Redis, Elasticsearch, Kafka and RabbitMQ, deployed on AWS with Docker, Kubernetes and Terraform. I care about making services fast and predictable under real load — caching strategy, query optimisation, retry and failover design. Open to senior backend engineering roles in Bengaluru or remote.

Location

Bengaluru

AI stack

antigravity

claude code

cursor

hugging face

ChatGPT

Work experience

Backend Software Engineer at Open Climate Fix, Jan 2026 - Oct 2026 · 10 mos, Bengaluru, Karnataka, India, * Developed a forecast-serving API in Python, from design through deploys. It combines several upstream providers and fails over to healthy ones during outages, so clients get a slightly less accurate forecast instead of an error. Live in three regions. * Merged 3 forked per-country codebases into one config-driven Python service, so launching a new country is a config change instead of a new codebase. * Cut infra spend 30% by moving 8 Python services off direct database access onto a gRPC API layer with generated clients, giving every service one source of truth. * Made the platform multi-tenant by moving API authorisation from per-site grants to organisation scope carried in the JWT, so one token covers every site an org owns. * Built Hexatron, a Go admin app for all 30 of our Airflow jobs, so the team triggers runs and reads logs themselves instead of pulling in an engineer. * Found records silently dropped on re-runs: dedup was keyed on timestamp alone, so once the national record saved, all 10 regional records at that timestamp were lost. Fixed the key and pinned it with a regression test. * Found that a rate-limit config that passed locally exempted every /v1 route in production, and set explicit limits on all 16 endpoints.

Software Engineer (Backend) at Orange Health Labs, Feb 2025 - Jan 2026 · 1 yr, Bengaluru, Karnataka, India, * Built and optimized backend services and REST APIs using Python and Go, improving reliability and performance across core healthcare workflows. * Reduced payout-related errors by 98% by decoupling earning and payout logic into independent services, backed by an AWS SQS worker with automatic retries. * Built JWT-based RBAC for Partner Invoices, implementing role-based access controls to protect sensitive financial data. * Automated monthly partnership invoicing using a Go scheduler, with a pre-check “dry run” workflow that reduced invoice errors by 95%. *Migrated order search from AWS RDS to Elasticsearch, reducing database load and improving search performance by 96%. * Developed backend functionality for correcting patient details before sample collection, reducing downstream data-quality issues. * Designed and worked with microservices, asynchronous processing, RabbitMQ, AWS SQS, Redis, Elasticsearch and relational databases.

Software Engineer (Backend) at Kaleyra, Jul 2022 - Jan 2025 · 2 yrs 7 mos, Bengaluru, Karnataka, India · Hybrid, * Built an HTTP service using FastAPI to handle thousands of requests per second, improving chatbot latency by 33%. * Designed and implemented a Python microservice for logging conversations to MongoDB via RMQ, enhancing system performance by 20%. * Optimised MongoDB queries and implemented sharding, improving search query speeds by 18%. * Reduced chatbot latency from 1.5 seconds to under 1 second using Redis Elastic Cache, boosting overall performance. * Re-engineered chatbot architecture for an Active-Active setup, achieving cross-region compatibility and eliminating single-region outages as a source of downtime * Built a retry mechanism that recovered 5% of requests that would otherwise have been dropped during peak traffic, keeping the service within SLA.

Software Engineer Intern at Kaleyra, Apr 2022 - Jul 2022 · 4 mos, Bengaluru, Karnataka, India · Hybrid, * Acquired hands-on experience in Full Stack Development. * Developed 4 fully functional web applications using the MERN Stack (MongoDB, Express.js, React, Node.js). * Implemented RESTful APIs and optimized database queries, resulting in a 20% improvement in data retrieval speed. * Integrated third-party services and APIs, reducing external dependency issues by 15%. * Conducted 3 peer code reviews to ensure code quality and best practices, improving team efficiency. * Led the deployment of applications on cloud platforms, enhancing scalability and reliability by 30%. * Received 40+ hours of soft skills training from certified trainers, improving communication and teamwork abilities.

Education

Bachelor of Engineering - BE, Computer Science, New Horizon College of Engineering

High School Diploma, Computer Science, KLE's CIM polytechnic hubli

Public links

github: https://github.com/PavanRaghavendraKulkarni

pavankulkarni-portfolio.netlify.app: https://pavankulkarni-portfolio.netlify.app/