OS
OMKAR_SAWANT
AI/ML Engineer & Automation Environment
INITIALIZING AI/ML CORE SYSTEMS...0%
Telemetry Log Ready
Python 3.12 • LangGraph • PyTorch
AI/ML ENGINEER • AUTOMATION BUILDER
A passionate developer who
Lovesbuilding&solvingproblems...
“If you can't buy one...★ then make one.🤠”
I'm a
Ibuildintelligentsystems,AI-poweredapplications,andzero-touchautomationworkflowsthatconvertcomplexdatachallengesintoreliablesoftwaresolutions.
BE Student atFAMT, Ratnagiri 🎓
Python
FastAPI
Docker
LangGraph
PostgreSQL
n8n
Git
Projects Built
Technologies
Public Repositories
Years Learning
Engineering Snapshot
Available for Internships & Projects
Primary Focus
AI/ML Engineering & Intelligent Automations
Core Stack
Python • FastAPI • LangChain • LangGraph • n8n • Docker
Education
FAMT, Ratnagiri (BE Student)
Open To
AI/ML Internships & Engineering Collaborations
Who I Am & What I Solve
Driven by Engineering Precision & AI Innovation
Combining machine learning models, Python backend logic, and zero-touch automations to build real working software.
Omkar Sawant
IamastudentdeveloperatFinolexAcademyofManagementandTechnology(FAMT),Ratnagiri.Ispecializeinturningcomplexproblemsintointelligent,reliablesoftwareapplications.
MycorefocusspansPythonbackenddevelopment,NLP,LLM-poweredapplications,andzero-touchn8nautomationpipelines.Itakeprideinclearsystemarchitectureandpracticalexecution.
Currently Exploring Roadmap
Agentic AI & LangGraphMulti-agent graph topologies and stateful tool execution
RAG & Vector SearchPinecone vector indexing and hybrid semantic retrieval
n8n Workflow EngineeringZero-touch document pipelines and automated Webhooks
LLM Fine-Tuning & EvaluationStructured outputs with Pydantic & Google AI Studio
Education & Journey
Bachelor of Engineering (Student)
CurrentFinolex Academy of Management and Technology (FAMT), Ratnagiri
FocusingonAI/MLalgorithms,Pythonsoftwaredevelopment,NLPtechniques,andintelligentsystemarchitectures.
• Data Structures & Algorithms in Python• Machine Learning & Neural Network Fundamentals• Campus FAQ Bot & Academic Assistant Projects
Automation & Workflow Systems Builder
Activen8n Prebuilt Stack & Open Source Ecosystem
Designingzero-touchautomatedbillingpipelines,multi-agentcontentcreationworkflows,andvectorsearchapplications.
• Multi-Agent Content Orchestration• Zero-Touch Invoice Document Parsing• Ollama + Pinecone RAG Automations
AI Research & Future Focus
OngoingLLM Applications & Industry Collaborations
ActivelyseekingAI/MLengineeringinternships,researchopportunities,andopen-sourcecontributionstoproductionAItools.
• Autonomous AI Agent Architectures• Enterprise Workflow Automation• Production LLM System Safety & Speed
Open for AI/ML Internships & Collaborations
Engineered Stack
Technical Stack & Ecosystem
Categorized technologies organized by application depth—no fake percentage bars or hype.
AI / ML & Data Science
NLP & LLMs(Core)
Scikit-Learn(Core)
NumPy & pandas(Core)
Matplotlib(Frequently Used)
Google AI Studio(Frequently Used)
PyTorch Basics(Exploring)
RAG Evaluation(Exploring)
Programming Languages
Python(Core)
JavaScript(Frequently Used)
HTML5 & CSS3(Frequently Used)
SQL(Core)
Automation & Agents
n8n Workflows(Core)
LangChain(Core)
LangGraph(Core)
Google Opal(Exploring)
Backend & Services
FastAPI(Core)
Flask(Frequently Used)
Pydantic(Core)
Streamlit(Core)
Docker(Frequently Used)
REST APIs(Core)
Databases & Tools
PostgreSQL(Core)
Pinecone(Frequently Used)
SQLite(Frequently Used)
Git & GitHub(Core)
Linux / Bash(Frequently Used)
Engineering Methodology
How I Build Systems
A systematic 5-step approach to turning complex problems into reliable, automated software.
01
UNDERSTAND
Deconstructcoreproblemspecificationsandsystemconstraints.
Structuredataflowsandtargetstatespecifications.
Problem Definition
02
DESIGN
Architectdatapipelinesandmodelinteractionlogic.
StructureAPIendpoints,vectorindexes,andprompting.
Architecture
03
BUILD
Implementhigh-performancePythonbackendsandAPIs.
LeverageFastAPI,Streamlit,LangChain,andvalidation.
Engineering
04
AUTOMATE
Eliminatemanualtaskswithzero-touchworkflows.
Connectn8nnodes,documentOCR,andagentpipelines.
Automation Stack
05
ITERATE
Benchmarkspeed,optimizeprompts,andtestedgecases.
Continuousevaluationtokeepsystemsfastandreliable.
Refinement
Selected Work
Systems I've Built & Deployed
Case studies detailing the real engineering problems, architectural choices, and practical outcomes.
Flagship Case Study
AI SYSTEM / ATS ANALYSISCompleted
SkillBridge – AI Resume Analysis & Guidance Bot
AI-powered resume evaluation & ATS screening guidance tool.
The Problem
JobapplicantsfrequentlyfaceautomatedATSrejectionswithoutfeedbackonwhytheirresumefailedorwhatspecificskillstheyneedtolearnnext.
The Solution
Engineeredanintelligentresumeevaluationenginethatparsescandidateresumes,mapsskillsagainsttargetedrolerequirements,flagsATSparsingissues,andgeneratesastructuredlearningroadmap.
Engineering Approach:
Python,NLPtextparsing,LLMprompting(Gemini&OpenAIAPIs),Pydanticoutputschemavalidation,StreamlitWebGUI.
Capability / Outcome:
Generatesactionableline-by-lineATSimprovementstepsandtargetedskillgaproadmapsinseconds.
PythonNLPLLMsATS ScreeningStreamlitPydantic
VISUAL AI TUTOR / RAGCompleted
MindFlow – AI Visual Tutor
Converts YouTube video lectures into interactive visual diagrams and knowledge nodes.
The Problem
Studentsandself-learnerslosehourswatchinglengthytechnicallectureswithoutclearvisualsummariesorinstantinteractiveQ&Acapabilities.
The Solution
DevelopedanAIvisualtutorpipelinethatautomaticallyextractsYouTubetranscripts,synthesizeskeyconcepts,andconvertscomplexideasintodynamicMermaid.jsflowchartsandinteractiveexplanations.
Engineering Approach:
Python,YouTubeTranscriptAPI,FastAPIbackend,Mermaid.jsgraphgeneration,LLMcontextintegration.
Capability / Outcome:
Converts60+minutetechnicalvideosintostructured,visualknowledgeflowchartsinstantly.
PythonFastAPIYouTube APIMermaid.jsLLMs
AUTOMATION PIPELINECompleted
Zero-Touch AI Invoice Processing
End-to-end automated document ingestion and validated structured data pipeline.
The Problem
Manualinvoiceingestionandmanualbillingdataentryiserror-prone,labor-intensive,andslowforfinanceworkflows.
The Solution
EngineeredanautomateddocumentvisionandOCRpipelinethatingestsrawPDFinvoices,extractsstructuredlineitems,validatesmathtotals,andformatscleanJSON/databaseoutputs.
Engineering Approach:
Python,FastAPI,Vision&OCRmodels,Dockercontainerization,schemavalidationrules.
Capability / Outcome:
Zero-touchautomatedinvoiceprocessingreducingmanualdataentryworkbyover90%.
PythonFastAPIOCR & VisionDockerAutomation
MULTI-AGENT SYSTEMOpen Source
Multi-Agent AI Workflow for Content Creation
Autonomous multi-agent system coordinating research, drafting, and post formatting.
The Problem
Consistentlycreatinghigh-qualitytechnicalcontentrequiresrepetitivemanualstepsacrossresearch,drafting,editing,andpromptgeneration.
The Solution
Designedamulti-agentorchestrationgraphwherespecializedAIagentscollaborate—agent1researchestopics,agent2writespostdrafts,agent3refinestone,andagent4generatesvisualprompts.
Engineering Approach:
LangChain,LangGraphstatefulmulti-agentsystem,n8nwebhookworkflowautomation.
Capability / Outcome:
Reducestechnicalcontentcreationtimefromhourstominutesviacoordinatedagentexecution.
LangChainLangGraphMulti-Agentn8n Workflows
Python///
FastAPI///
Docker///
LangGraph///
n8n Automation///
PostgreSQL///
LangChain///
Git & GitHub///
Python///
FastAPI///
Docker///
LangGraph///
n8n Automation///
PostgreSQL///
LangChain///
Git & GitHub///
Python///
FastAPI///
Docker///
LangGraph///
n8n Automation///
PostgreSQL///
LangChain///
Git & GitHub///
Open Source Lab
Repositories & Public Tools
Explore open-source repositories, automations, and backend tools built across GitHub.