Open to remote roles · Nairobi, Kenya
Kennedy Kinyanjui
BackendEngineer
Backend and Data engineer in Nairobi, building the microservices, pipelines and dashboards behind business-support systems at telecom scale. I came up through site reliability, so I design for the 3 a.m. pager as much as for the happy path.
- 30k+
- req/s peak traffic handled
- 94%
- test coverage on core service
- 2h → 10m
- MTTR on routed incidents
About
How I work
I'm a full-stack engineer currently working at Safaricom PLC where I build and own the services that keep business-support processes running for one of Africa's largest telcos — Java/Spring Boot and FastAPI on OpenShift, Kafka Apache spark Airflow Postgres,MongoDB and Oracle underneath, React and TypeScript on top.
I got here through site reliability engineering, and it shows in how I build. I care about idempotent jobs, honest observability, and failure modes you can reason about at 3 a.m. — because for a few years, 3 a.m. was when I found out whether I'd been right.
Outside work I build things to understand them: computer vision pipelines, streaming data architectures, MEV bots on Polygon, and an algorithmic trading system that runs unattended on my own VPS. I write about what I learn on dev.to.
Currently
- Building BSS automation services at Safaricom PLC
- Writing about AI, data engineering and SRE on dev.to
- Working toward an AWS AI/ML certification
- Open to remote engineering roles
Work
Selected projects
01 · 2026Private
AI Ticket Classifier Bot
An agent that reads incoming incident email, classifies it, resolves the owning team against the live on-call directory, and files the ticket automatically. LLM classification over Microsoft Graph, with a confidence threshold that routes anything ambiguous to a visible human queue rather than guessing. Cut mean time to resolution on routed incidents from roughly two hours to about ten minutes.
02 · 2025Private
Peak-Traffic Autoscaling & DR Failover
Telemetry-driven automation from my SRE years: Ansible and Bash playbooks fed by Dynatrace and ELK signals that autoscaled OpenShift workloads and failed traffic over to DR sites during peak windows — keeping CRM, CVM and mobile-money transaction paths available at sustained peaks of 20,000–30,000+ requests per second.
Experience
Where I've built
Software Engineer — BSS Automation / RPA Squad
TODO – PresentSafaricom PLC
Backend and full-stack delivery for business-support automation across credit control, line recycling and outbound campaigns.
- Own backend microservices in Java/Spring Boot and FastAPI running on OpenShift, backed by MongoDB and Oracle
- Diagnosed and fixed a cache-key defect causing 99.9% silent record loss in a production data pipeline, then took test coverage to 94% with JaCoCo
- Shipped an AI ticket-classification agent that reduced MTTR on routed incidents from ~2 hours to ~10 minutes
- Build React/TypeScript operational dashboards with Entra ID SSO and Azure AD-driven access control
- Author high-level design documents for platform-level projects
Site Reliability Engineer
TODO – TODOSafaricom PLC
Availability and automation for customer-facing transaction paths at national scale.
- Kept CRM, CVM and mobile-money transaction paths available at sustained peaks of 20,000–30,000+ requests per second
- Built telemetry-driven automation that autoscaled OpenShift workloads and performed DR site failover during peak windows
- Ran workflow and order-management orchestration on Temporal and Camunda BPM
- Owned observability tooling across Dynatrace and the ELK stack
Software Developer
TODO – TODOAfristec (e-kraal)
Full-stack product work on a government communications platform.
- Built Node.js microservices for Talking Drum, an internal chat, ads and video platform for Kenyan government departments
- Delivered features across both the service and client layers
Toolkit
Skills & education
Languages
- Java
- Python
- TypeScript
- JavaScript
- SQL
- Bash
- Solidity
Backend
- Spring Boot
- FastAPI
- Node.js
- Express
- REST
- Microservices
- Async Python
Frontend
- React
- Next.js
- Vite
- Tailwind CSS
- Recharts
Data
- MongoDB
- Oracle DB
- PostgreSQL
- Redis
- Apache Airflow
- PySpark
- Kafka
Platform & SRE
- Kubernetes
- OpenShift
- Docker
- Jenkins
- Ansible
- Dynatrace
- ELK
- AWS
Orchestration & AI
- Temporal
- Camunda BPM
- LLM integration
- RAG
- YOLO
- scikit-learn
Education
Bsc. Computer Science
Moi University, Eldoret ,Kenya
2013-2017
Certifications
- AWS Certified — Cloud Practitioner, AI practioner
- AXELOS certified- ITILV4 service management suite
Contact
Let's build something that stays up.
I'm open to remote backend, full-stack and platform roles, and always happy to talk about distributed systems, data pipelines, or what broke in production last week.