Senior Backend & AI Engineer - 7+ yrs - Distributed Systems, LLM
Resume:
ASHENAFI MULUGETA
ashumulugeta*** @gmail.com
Addis Ababa, Ethiopia
SUMMARY
Senior Backend & AI Engineer with 7 years of experience architecting high-performance distributed systems and production-grade LLM orchestration. Proven track record of scaling architectures to handle 10M+ monthly requests and building complex agentic workflows using GPT-4 and LangChain. Expert in Python, Go, and event-driven microservices (Kafka) with a deep focus on system reliability, security, and data integrity. Adept at navigating
ambiguous requirements, documenting architectural reasoning, and mentoring engineering teams to deliver elite-level software.
PROFESSIONAL EXPERIENCE (BACKEND & AI ENGINEERING)
Upwork, Senior AI/ML Engineer 05/2021 – 08/2025 remote
Built 12+ production AI systems including fraud detection (Kafka + XGBoost, 5M+ daily transactions), document OCR, and LLM chatbots using LangChain and GPT-4.
Led a 4-person team to develop customer support automation that reduced response time by 70%. Mentored 6 junior engineers on model versioning, A/ B testing, and automated retraining.
Designed multi-step agentic workflows requiring tool use,reasoning documentation, and handling under specified client requirements.
Agile Lab, Senior Backend & Data Enginee 07/2022 – 08/2024 Italy remote
Architected and scaled a high-concurrency backend ecosystem handling 10M+ monthly requests with sub-100ms latency, leading the transition from a legacy monolith to event-driven microservices (FastAPI, Go, Kafka) that increased system throughput by 3x.
Engineered a secure multi-tenant SaaS architecture for 500+ enterprise clients, implementing PostgreSQL row-level security (RLS) and robust data isolation protocols to meet stringent global compliance and data integrity standards.
Directed technical excellence and team growth by mentoring 4 engineers on advanced system design and code quality, while establishing CI/CD and observability patterns that reduced deployment errors and improved architectural consistency across the organization.
Software Engineer, Fintech & Freelance 09/2016 – 05/2018
Built internal ERP modules and payment gateway integrations for a local financial institution.
Developed small business websites, dashboards, and basic automation scripts.
AI TRAINING & LLM EXPERIENCE Mindrift, AI Annotator & QA Lead
12/2023 – 06/2025 remote
Evaluated LLM responses and search engine results for safety, helpfulness, and factuality. Reviewed submissions from 50+ annotators and provided constructive feedback to improve quality by 35%.
Designed annotation guidelines for ambiguous, multi-step reasoning tasks. Documented both actions taken and reasoning behind decisions directly relevant to OpenClaw trajectory validation.
Machine Learning Research Assistant, Addis Ababa Science and Technology University 01/2022 – 05/2023 Addis Ababa Ethiopia
Implemented PyTorch training pipelines for solar cell defect segmentation using Mask2Former. Automated data preprocessing and experiment tracking with MLflow to standardize reproducibility.
Focused on research workflows requiring careful documentation, defensible methodology decisions, and multi-step reasoning.
•PROJECTS
LearnWise - Adaptive Learning Platform with LLM Assistant, FastAPI, Cassandra, Kafka, PyTorch, LangChain,GPT-4, Azure Kubernetes 06/2023 – 06/2025
Built recommendation engine using collaborative filtering and BERT for personalized learning paths serving 2M+ users. Architected real-time progress pipeline processing 100k events/day via Kafka streams.
Integrated LLM-based tutoring assistant (GPT-4 + RAG) deployed on Azure Kubernetes. Handled ambiguous user queries requiring multi-step reasoning and tool use.
Led a team of 5 engineers; conducted code reviews and architecture design sessions.
PayStream – High-Throughput Payment Orchestration, Go, Python, FastAPI, Kafka, Redis, AWS ECS, Terraform 01/2022 – 02/2023
Built payment orchestrator routing to 5+ gateways with automatic failover and 99.99% uptime. Implemented idempotency, exponential backoff retries, and dead-letter queues.
Added real-time fraud detection (rules + ML) reducing chargebacks by 18%. Designed multi-step transaction workflows with clear logging and audit trails.
Study Buddy – Real-Time Collaborative Platform, Django, React, WebSockets, Redis, Docker, PostgreSQL 09/2020 – 08/2021
Built full-stack platform with WebSocket chat, Pomodoro timers, Jitsi meetings, and streak analytics. Integrated Chapa payments, JWT + Google OAuth, API throttling, and caching.
Containerized with Docker Compose (PostgreSQL, Redis, Celery, Nginx). Led architecture decisions and mentored one junior developer.
Demonstrated ability to integrate multiple tools (payment, auth, real-time, video) into coherent workflows.
EDUCATION
Addis Ababa Science and Technology University, Software Engineering 03/2014 – 08/2019 Addis Ababa, Ethiopia
Built a strong foundation in distributed systems, machine learning, algorithms, database systems, and system design.
Applied theoretical concepts to practical projects involving backend development and data-driven applications.
Coding Academy, Advanced Algorithms & Data Structures
04/2017 – 06/2019 Addis Ababa, Ethiopia
Completed an intensive program focused on advanced data structures, algorithmic problem solving, and system optimization.
Solved 700+ algorithmic problems on platforms such as LeetCode and Codeforces, demonstrating strong analyticaland problem-solving skills.
Strengthened expertise in writing efficient, scalable, and optimized code for real-world applications.
SKILLS
Languages — Python, TypeScript, Go(Golang), Java, SQL, JavaScript, Bash
Frontend & Backend — React, FastAPI, Django, Node.js, Celery, Redis, Kafka, PostgreSQL
Machine Learning & AI — PyTorch, TensorFlow, Scikit-learn, LangChain, ONNX Runtime, OpenCV, Hugging Face Transformers, Azure Cognitive Services, Model Deployment & Optimization
Tools & Infrastructure — Docker, Kubernetes, AWS, Git, Linux, CI/CD Pipelines, Jupyter Notebooks
LLM & AI Evaluation — Prompt Engineering, RLHF, Data Annotation, Annotation Guideline Design, Model Evaluation, Factual Accuracy QA, Response Ranking & Benchmarking
Contact this candidate