Uzma Saajid

Backend Engineer

Maharashtra, India4 yrs 11 mos experience

Most Likely To SwitchAI Enabled

Key Highlights

Expert in building scalable AI platforms.
Proficient in LLMs and RAG pipelines.
Strong background in data engineering and backend architecture.

Stackforce AI infers this person is a SaaS-focused AI/ML Engineer with strong data engineering capabilities.

Contact

Skills

Core Skills

Large Language Models (llm)Retrieval-augmented Generation (rag)Etl

Other Skills

Adobe Creative CloudAirflowAndroid DevelopmentAndroid StudioApache AirflowApache KafkaApache Spark StreamingAsyncIOAzure DevOpsBig Data AnalyticsCloud ComputingComputer VisionData AnalysisData ScienceDeep Learning

About

AI/ML Engineer and Data Scientist with 4+ years of experience designing and deploying intelligent data-driven systems, combining strengths in machine learning, backend architecture, and data engineering. Proficient in LLMS, RAG pipelines, and NLP, with hands-on experience building scalable, cloud-native Al platforms and automating data pipelines. Adept at bridging data science experimentation with production-level implementation to deliver reliable, high-performance Al solutions.

Experience

4 yrs 11 mos

Total Experience

2 yrs 5 mos

Average Tenure

3 yrs 6 mos

Current Experience

Codvo.ai

3 roles

Senior Backend Engineer - I

Promoted

May 2024 – Present · 2 yrs · Remote

Enterprise GenAl & Document Intelligence Platform (NelO.ai)
Architected enterprise RAG pipeline for 10K documents achieving <3 s query latency and 97% contextual accuracy.
Developed Al skill modules (summarization, classification, fact extraction, reconciliation) using LLMs (OpenAl, Gemini)
Built Text-to-SQL engine enabling natural-language queries via Haystack, PostgreSQL, and embeddings.
Designed semantic indexing system using Qdrant and OpenAl embeddings for document retrieval and context linking.
Implemented microservices architecture with FastAPI, NATS, and Azure DevOps for scalable Al processing.
Added observability (OpenTelemetry) and distributed tracing to monitor Al job pipelines.
Created multi-database connectors (PostgreSQL, Oracle, MySQL, SQL Server) for unified access.
Improved Al throughput by 0% and reduced latency by 60% through optimized async processing and vector search.
Sub-Batch Document Processing System
Developed distributed OCR + text extraction pipeline using Gemini Al, Surya OCR, PyTorch, and AsyncIO.
Automated document-to-markdown conversion and error recovery using NATS JetStream
NelO Backend & Skill Master Services
Built 100+ REST APIs for document insights, Al job management, and multi-tenant access.
Integrated Keycloak RBAC, Azure Blob storage, and NATS event streaming.
Orchestrated review workflows, rerun versioning, and job state tracking with telemetry logging.

Large Language Models (LLM)Retrieval-Augmented Generation (RAG)FastAPINATSAzure DevOpsOpenTelemetry+5

Backend Engineer - III

Jan 2024 – Jun 2024 · 5 mos · Remote

Backend Engineer - II

Nov 2022 – Jan 2024 · 1 yr 2 mos · Remote

Siemens healthineers

Design And Development Engineer

Jun 2021 – Nov 2022 · 1 yr 5 mos · Bangalore Urban, Karnataka, India

Designed and developed ETL data integration pipelines to ingest and transform healthcare equipment data from Oracle into Teradata and Snowflake using SAS DI Studio,
Built automated Data Quality Framework in Python & SQL for data validation, incremental loads, and metadata consistency.
Created Power BI dashboards for real-time quality metrics and performance tracking.
Implemented Airflow orchestration and CI/CD pipelines using Azure DevOps.
Improved pipeline efficiency and reduced manual testing overhead by 30%.

ETLPythonSQLPower BIAirflowAzure DevOps