Experience — Ayush Mandowara's Career Journey
Over 7+ years of building and scaling AI/ML systems across startups and enterprises. From hands-on ML engineering to technical project management, my journey spans the full spectrum of AI product development — from ideation and prototyping to production deployment and continuous optimization.
[Work]
Virtusa
Senior GenAI Consultant (client: United Health Group)
Aug 2025 – Present
Gurgaon, India
- • Leading development of Advocate Assist — an enterprise-grade conversational AI agent for proactive healthcare support, built on Google Cloud Platform using Gemini 2.0, Dialogflow CX, and Playbooks.
- • Architecting production-grade AI systems handling 10K–100K+ calls/day with sub-second latency, serving thousands of healthcare advocates and members.
- • Collaborating directly with solution architects, product owners, delivery managers, and UHG business stakeholders to define AI strategy, delivery roadmaps, and success metrics.
- • Driving best practices in prompt engineering, RAG pipeline design, evaluation frameworks, and responsible AI governance for healthcare-domain applications.
GenAIGemini 2.0Dialogflow CXGCPPythonRAG
[Work]
adeptmind.ai
ML GenAI Engineer → Technical Project Manager - DLP Team
Sep 2023 – May 2025
Remote, Delhi, India
- • Built the high-revenue Dynamic Landing Pages (DLP) product from scratch under CTO guidance, leading ML/GenAI efforts to auto-generate SEO-optimized retail pages that improved Google rankings and boosted client footfall.
- • Built, scaled and maintained an in-house LLM service handling 100K+ requests/day using FastAPI, Celery, and Redis.
- • Led cross-functional planning (Scrum, retrospectives), interfacing with product, engineering, and business stakeholders to align roadmap and delivery.
- • Mentored new hires, contributed to core model development and MLOps (GitHub Actions, Kubernetes, Prefect) and drove documentation and process maturity.
FastAPICeleryRedisLLMKubernetesGitHub ActionsPrefect
[Work]
adeptmind.ai
Machine Learning (Natural Language Processing) Engineer - Search Team
Apr 2022 – Sep 2023
Remote, Toronto, Canada
- • Maintained Content Understand (CU) and Query Understanding (QU) Machine Learning components using PyTorch, HuggingFace, FastAPI, distilbert, nlp.
- • Built and deployed category & classification pipeline via AWS Sagemaker; deprecated Tensorflow.
- • Built scraping tools and live inventory software (Python, Go, TypeScript, SQL) for malls, retailers.
PyTorchHuggingFaceFastAPIAWS Sagemakerdistilbert
[Work]
now.gg / BlueStacks.com
Senior Member of Technical Staff – QA Automation Team
Sep 2021 – Apr 2022
Remote, Gurugram, India
- • Performance optimization for now.gg BlueStacks Android App Player on Cloud.
- • 5X reduction in test development time and 2X reduction in Suite Run Time using Data Analysis.
- • Regression models for revenue prediction & release funnel tracking (big data, p-value analysis).
PythonData AnalysisBig Data
[Work]
BlueStacks.com
Member of Technical Staff – Technical Support → QA Automation Team
Jun 2019 – Aug 2021
Remote/Onsite, Gurugram, India
- • Developed a multi-lingual PyTest-based automation suite for BlueStacks 5 (Windows 7, 8, 8.1, 10).
- • Developed NLP-powered tools over Zendesk for SEO, log analysis, and automated replies.
- • Integrated and customized Kommunicate.io – DialogFlow live chat into Zendesk and App Player.
PyTestNLPDialogFlow