I am a GenAI Engineer with over 3 years of experience transforming advanced AI concepts into production-ready solutions that reduce costs, speed delivery, and boost accuracy. My expertise lies in building safe and observable AI workflows that deliver measurable ROI and earn trust from executives and stakeholders through strong end-user adoption. I excel at collaborating across functions, translating deep technical skills into impactful business outcomes that resonate with clients and leadership.
Throughout my career, I have optimized multi-agent workflows and AI infrastructure to significantly improve latency, throughput, and operational costs at scale. I have developed robust MCP servers with audit trails and orchestration tools that reduce integration lead times and enforce safety guardrails, ensuring zero PII incidents in high-volume environments.
I have launched Agentic RAG systems that enhance accuracy and groundedness while deflecting support tickets, saving operational expenses. My work includes creating LLM-driven pipelines and analytics copilots that reduce forecast errors, automate workflows, and improve data integrity with AI observability tools.
I am skilled in deploying scalable machine learning models using TensorFlow, Docker, and CI/CD pipelines, improving inference latency and experiment cycle times. I have experience building microservices for batch inference and benchmarking, which streamline analyst reviews and improve model deployment.
My projects include developing multi-agent travel planners, autonomous delivery agents, and commerce copilots that leverage hybrid retrieval, policy-aware RAG, and audit trails to deliver high accuracy, cost savings, and user-friendly automation. I am passionate about advancing AI infrastructure and observability to create safe, efficient, and impactful AI systems.
Optimized vLLM+KV-cache behind LangServe and redesigned multi-agent workflows, cutting P95 latency 45%, doubling throughput, and reducing task cost 33% at production scale. Developed MCP servers with audit trails and n8n orchestration, reducing integration lead time from 3 weeks to 3 days with zero PII incidents across 60k+ tool calls; enforced safety guardrails. Launched Agentic RAG, improving groundedness +12 pts, accuracy +9 pts, and deflecting 22% of tickets, saving $3.8k/month in OpEx.
Created LLM-driven pipelines with LangGraph and LangSmith tracing, reducing forecast error 18%, false positives 27%, and boosting precision +11 pts. Shipped NLβSQL analytics copilot automating Power BI workflows, eliminating 12 hrs/week manual ELT, cutting time-to-insight 55%, and increasing adoption +35%. Implemented Great Expectations, AI observability, and LlamaIndex RAG assistant, improving data integrity +23% and cutting incidents 60% while maintaining SLA 99.7%.
Boosted F1 score 0.71β0.83 and reduced inference latency 35% using TensorFlow pipelines, transfer learning, augmentation, and quantization. Scaled dataset 5x and reduced experiment cycle time 7β3 days with reproducible MLflow workflows and parameterized training scripts. Packaged models as Dockerized Flask microservices with batch inference and CI benchmarking, reducing analyst review time 30% across 12 variants.
Owned SQL pipelines with validation checks, reducing prep time 58% and stabilizing refresh reliability to 99.4% for flagship product. Built scikit-learn baselines with targeted feature engineering, improving AUC 0.74β0.82 and enabling earlier case detection. Delivered KPI dashboards for clinical stakeholders, shortening decision lead-time by 3 days and enabling 3 new research cohorts.
Jobicy
578 professionals pay to access exclusive and experimental features on Jobicy
Free
USD $0/month
For people just getting started
Plus
USD $8/month
Everything in Free, and: