EI
Sr Python Developer
Echo IT Solutions, Inc
- 🇺🇸 United States
- On-site
- Senior
- 7 hours ago
- Bedrock
- Server-Sent Events
- Elasticsearch
- Aurora
- PostgreSQL
- pgvector
- Windows
- Python
- Azure Synapse
- Snowflake
- Machine Learning
- ETL
- AWS Glue
- Apache Airflow
- Pandas
- KYC
- GPT
- MLOps
- DynamoDB
- Java
- Microservices
- React.js
- REST API
- Flask
- FastAPI
- Django
- Node.js
7 hours ago
- Designed and led delivery of aBedrock-based agentic support platform— tool-calling agents over a curated skill library with SSE streaming — that now handles40M+ requests/month and deflects 61% of tier-1 support tickets, reducing median time-to-resolution from 34 minutes to 7
- Designed the skill library abstraction — versioned, permission-scoped troubleshooting capabilities the agent composes at runtime — allowing new diagnostic workflows to ship without model or prompt-layer changes.
- ConfiguredBedrock Guardrails and Knowledge Bases to enforce tenant-scoped grounding and output constraints, ensuring one tenant's telemetry and runbooks could never surface in another's session across a ~80-tenant fleet.
- Owned provisioned throughput sizing and inference cost tuning, modeling peak concurrency against committed capacity to hold unit cost predictable as the fleet scaled.
- Built the retrieval layer as hybrid search over self-managedElasticsearch (BM25) andAurora PostgreSQL,pgvector(dense), tuning chunking and re-ranking against a labeled relevance set rather than eyeballed examples.
- Cut generative inference spend 58% ($2.1M annualized) by introducing Model-Tiering (routing 78% of traffic to smaller models via a learned classifier), prompt caching, and reserved provisioned throughput for peak windows — without measurable regression on the eval suite.
- Architected and managed data warehousing solutions usingPython, Azure Synapse Analytics andSnowflake, optimizing data partitioning, indexing and transformation pipelines for high-performance querying
- Monitored Snowflake performance and query optimization using Python scripts and Snowflake's INFORMATION_SCHEMA views.
- Designed and implemented GenAI-powered ETL pipelines usingAWS Glue, Apache Airflow andPandas, enabling automated KYC validation, risk profiling and suspicious transaction monitoring (CTR & SAR submissions to FINCEN) by leveragingGPT-based document parsing.
- ImplementedMLOps pipelines integratingDynamoDB and S3 for scalablefeature engineering and training datasets.
- Designed and integratedAgentic AI frameworks with Java-based microservices to enable autonomous task execution, decision-making, and workflow orchestration.
- IntegratedReact frontend with Python-based REST APIs (Flask/FastAPI/Django) for real-time data exchange.
- Created end-to-end test suites for Lang Graph workflows, validating correct node transitions, memory management, and stateful logic.
Sr Python Developer · Echo IT Solutions, Inc