Anant Patankar · India

ML/AI Engineer — Agentic Systems & Generative AI

Multi-agent orchestration (Google ADK) · MCP Protocol · Production RAG · MLOps at scale

Forensic scientist turned ML engineer — I bring evidence-grade rigor to agentic AI and grounded retrieval.

What I do

01

Agentic AI

Multi-agent orchestration with Google ADK, MCP protocol servers, and LLM-driven reasoning & tool use — systems where the model decides, and the architecture keeps it safe, cheap, and auditable.

02

Production RAG

Grounded generation with source traceability, hybrid dense+sparse retrieval, cross-encoder reranking — measured with real recall and latency numbers, not vibes.

03

Production Systems & MLOps

Async Python, distributed pipelines (Celery/Redis/Cloud Run), containerized deploys, experiment tracking and model registries — the backbone that makes the AI shippable.

Skills

Tech Stack

01 / Agentic AI & LLMs

Google ADKMCP ProtocolMulti-agent orchestrationLangChainLlamaIndexGeminiClaudeOpenAI APIAzure OpenAIHuggingFaceRAGReasoning & Tool Use

02 / Vector Search & Retrieval

FAISSQdrantHNSWLIBElasticsearchVertex AI SearchHybrid SearchCross-encoder RerankingSemantic Embeddings

03 / ML / Deep Learning

PyTorchTensorFlowscikit-learnTransformersNumPyPandasNLPTime Series

04 / Distributed Systems & Infra

Async Python / asyncioFastAPICeleryRedisRabbitMQDockerKubernetesCloud RunWebSockets / SSECI/CD

05 / MLOps & Cloud

MLflowDVCModel RegistryInference OptimizationGCP · Vertex AIAzure · OpenAI / CosmosDBAWS · S3 / SageMaker / Lambda

06 / Data Engineering

SQL / SQLglotPySparkETL PipelinesFeature EngineeringData ValidationSchema Management

Selected work

Featured Projects

Five systems spanning agentic engines, MCP servers, and RAG pipelines. Two are open source and independently verifiable.

FLAME — Agentic Recommendation Engine

Google ADK · Gemini 2.5 Flash · GCP · Flagship

AI backend for a gamified crowd-labelling platform that turns plant workers into a distributed workforce annotating workplace-safety imagery through four mini-games. Fully LLM-driven topic selection over structured Firestore tools — no hardcoded routing — plus consensus, Elo ranking, trust scoring, and anti-gaming.

12 agents · 5 ADK skills · designed for 100K+ concurrent users & 50K+ images/day

Google ADKGemini 2.5 FlashVertex AI SearchFirestoreCloud RunFastAPI

AI-Powered HSE Training Content Generation

Vertex AI Agent Engine · RAG · Human-in-the-loop

14-agent content-generation pipeline that turns safety documents into reviewed microlearning — grounded RAG with per-question source attribution and a structural guarantee that nothing publishes without human approval.

14 agents · multi-format RAG ingestion · zero auto-publish path

Vertex AI Agent EngineGemini 2.5 FlashVertex AI SearchFirestore

Lumenore Analytics MCP Server

MCP Protocol · Async Python
Open Source

Production MCP server exposing a BI platform’s analytics to any AI assistant — natural-language querying, forecasting, correlations, outlier detection — over typed, JWT-scoped, SSE-streaming tools.

sub-50ms P95 · engineered for 100+ concurrent connections · 8 analytics tools

FastMCPasyncio / aiohttpJWTSSEDocker

MLOps MCP Server

MCP Protocol · MLOps · Personal OSS
PyPI · Open Source

Open-source MCP server giving AI assistants tool-based access to MLOps workflows — experiment tracking, model registry, dataset drift detection, pipeline DAGs, lineage tracing — wrapping MLflow, DVC, and Git.

pip install mlops-mcp-server · v0.1.0 · MIT · two-tier tool registry

FastMCPMLflowDVCPandera / SciPy

Enterprise Knowledge Q&A

RAG · Hybrid Search · Evaluation

RAG system over enterprise document libraries: ChromaDB + Elasticsearch hybrid retrieval, cross-encoder reranking, and token-budgeted prompts with mandatory citations — turning ~15-minute manual document hunts into sub-minute answers in internal pilot use.

85%+ recall@5 on 200 labelled queries · hybrid retrieval lifted a 65–70% dense-only baseline

ChromaDBElasticsearchAzure OpenAILangChainCross-encoder

Verifiable proof

Open Source

Don’t take my word for it — read the code.

mlops-mcp-server

PyPI

MCP server for MLOps workflows: experiment tracking, model registry, KS-test drift detection, DAG cycle detection (Kahn’s algorithm), BFS lineage tracing with Mermaid visualizations.

$ pip install mlops-mcp-server

v0.1.0 · MIT · Python 3.11+

lumenore-mcp

Org repo

Production analytics MCP server I authored at Netlink — FastMCP, async Python, JWT auth, SSE streaming, zero data storage. Published open source under the Lumenore Platform org.

MIT · Python 3.13 · FastMCP · 8 analytics tools

Track record

Experience

AI Engineer — Xevyte Technologies
Apr 2026 – Present · Bengaluru

Architected the agentic AI backend (FLAME) for an enterprise HSE gamification platform and a 14-agent training-content generation pipeline — Google ADK, Vertex AI, Gemini, Cloud Run. Led technical direction for a junior engineer; presented vector-DB trade-off analysis to senior stakeholders.

Software Engineer — Netlink Software
Jul 2025 – Apr 2026 · Bhopal

Built the open-source Lumenore Analytics MCP Server; an AI customer-support chatbot with Qdrant hybrid search + cross-encoder reranking (85%+ retrieval accuracy, sub-200ms P95 at 10K+ daily messages); and an agentic MCP client/orchestrator managing multiple MCP servers across 4 transport protocols with RBAC.

Senior Associate / Associate Data Scientist — Affine Analytics
Jun 2022 – Jun 2025 · Bengaluru

ML/GenAI work on an Enterprise Knowledge Q&A RAG system (85%+ recall@5 on labelled queries); code-analysis AI processing 500K+ lines across 10+ languages; 30+ ML/DL solutions for a national rural-development bank on 10M+ records. 2× Game Changer of the Month (Sept & Dec 2023).

Junior Data Scientist — RPDS Innovations
Jan 2021 – May 2022 · Pune

Time-series forecasting (RF, SVM, LSTM) for chemical-reactor optimization with feature engineering and anomaly detection.

Python Developer — Miscos Technologies
Aug 2019 – Feb 2020 · Pune

ML forecasting pipeline (ARIMA, RF, LSTM) and a real-time object-detection system processing 15 FPS video with TensorFlow.

About

From forensic science to agentic AI

I hold an M.Sc. in Forensic Science and moved into machine learning through self-directed study, building toward production ML and AI over nearly six years. The analytical discipline transferred directly: I design agentic and RAG systems with an emphasis on auditability, grounding, and measurable outcomes.

My current focus is multi-agent orchestration (Google ADK), MCP-based tooling, and production RAG, supported by distributed-systems and MLOps practices — with two open-source MCP servers published publicly.

Get in touch

Contact

Open to conversations about agentic AI, GenAI systems, and ML engineering roles.