MK
Monika Kushwaha
Contact details — on requestProof of Work — on request
Experience
T
Founder & Lead AI Engineer
TrustLLM · Oct 2025 – Present
- Architected end-to-end AI reliability platform; reduced hallucination rates by ~40% through automated evaluation pipelines
- Built modular Python backend services handling multi-agent orchestration, tool calling, and autonomous execution pipelines
- Designed RAG systems with FAISS achieving measurable improvements in retrieval grounding and response consistency
- Integrated LangChain, LangGraph, FastAPI, Docker, and CI/CD pipelines delivering production-ready AI services
- Implemented LLM-as-Judge evaluation workflows; improved response quality scoring accuracy across test suites
- Developed human-in-the-loop validation workflows reducing AI governance risk for enterprise-grade deployments
I
AI Engineer
Independent Generative AI Systems · Jan 2024 – Present
- Built Python-based LLM applications with modular architecture; improved maintainability and reuse across projects
- Designed RAG pipelines using FAISS; reduced context hallucination and improved answer grounding measurably
- Developed multi-step agentic workflows with decision-based execution - shipped 3 end-to-end autonomous systems
- Orchestrated multi-agent systems with LangGraph; reduced workflow execution complexity via specialised agent decomposition
- Built automated evaluation and regression pipelines with PyTest; caught output regressions before production deployment
- Implemented LLM-as-Judge validation; enabled systematic quality scoring replacing high-effort manual review
- Integrated Langfuse observability; provided full trace visibility into prompt chains, latency, and output quality
- Designed human-in-the-loop checkpoints improving system trustworthiness for sensitive AI decision points
- Applied TrustLLM evaluation techniques; improved robustness scores and reduced hallucination failure rates
F
AI Automation Consultant / GenAI Systems Engineer
Freelancer + Outlier AI · Jan 2024 – Sep 2025
- Delivered 5+ agentic AI workflow systems for clients; reduced manual product lifecycle effort by ~60% via automation
- Architected multi-step reasoning pipelines integrating LLMs with Slack, Notion, and Google Docs via REST APIs and OAuth
- Built scalable RAG architectures with vector databases; improved retrieval accuracy and reduced irrelevant LLM responses
- Developed automated evaluation pipelines measuring grounding, hallucination detection, and semantic consistency
- Created reusable Agile automation templates adopted across sprint planning, retrospectives, and reporting workflows
- Contributed LLM quality assessment workflows to large-scale Outlier AI evaluation systems
C
Senior Test Engineer
Coforge Ltd. · May 2021 – Dec 2023
- Built Python automation frameworks reducing manual API regression effort by ~70% across enterprise backend systems
- Designed and validated REST APIs with OAuth and token-based auth; integrated with GitHub Actions and Azure DevOps CI/CD
- Developed AI/ML output validation systems ensuring reliability and consistency across production model deployments
- Collaborated with cross-functional teams in Agile/Scrum environments; consistently met sprint delivery targets
I
Quality Analyst
Intelenet Global Services · May 2016 – Jun 2017
- Executed end-to-end validation workflows improving release reliability for enterprise production systems
- Supported Agile/Kanban workflows ensuring on-time feature delivery and measurable quality improvements
Quality Analyst
Tech Mahindra · Apr 2015 – Apr 2016
- Developed regression testing frameworks stabilising enterprise applications across release cycles
- Collaborated with developers and business teams to validate requirements and accelerate release quality
Skills 0 proven through work
Also works with
Kano Model ApplicationString InterpolationData-Driven RecommendationsConsistent Quality DeliveryWorkflow Design (Process Mapping)Performance TestingMoSCoW PrioritizationRICE PrioritizationWork Planning and PrioritizationProject ExecutionStrategic PlanningSoftware Quality AssuranceBefore-and-After Impact MeasurementStability ImprovementGoal setting (defining key results)Low-Level System DesignApplication MonitoringEvaluation Framework DesignTask AutomationGenerative AI ApplicationData CurationDummy Data UsageTest Suite ManagementPytest AutomationCI/CD Pipeline ImplementationVersion Control with GitHubRegression TestingBatch Processing ImplementationAsynchronous ProgrammingAPI Authentication ImplementationAWS IAM & Access ManagementEmail Sending IntegrationNotification System DevelopmentSemantic SearchAWS Textract UsageMicroservices IntegrationAuthentication System IntegrationCustom Filter DevelopmentFeature EngineeringAmazon S3CSSHTMLVector Database UsageEmbedding GenerationSentence Transformer UsageText EmbeddingPostgreSQL Database UsageUnstructured Text Cleaning & PreparationOCR System DevelopmentREST API Endpoint DevelopmentNLP Application DevelopmentAI Workflow AutomationAI Model ConfigurationLoss Function OptimizationPrompt EngineeringTraining Data CreationTrain-Test Split ValidationHyperparameter TuningData Unification and NormalizationAnomaly DetectionLLM-as-a-Judge EvaluationRagas Framework UsageCosine SimilarityModel Metric Selection & InterpretationFunctional TestingCustom Metric DesignAPI Testing with PostmanFastAPI DevelopmentLangChain Framework UsagePDF RAG System DevelopmentBackend DevelopmentPython DevelopmentLLM Output Verification and ValidationAI Bias MitigationAI Hallucination DetectionAI Model TestingAI Platform Architecture Design
Proof of Work
Proof of Work
Monika shares this with people who ask. You'll hear back either way.
Education
B.Tech, Information Technology
Guru Jambheshwar University of Science & Technology · 2009 — 2013
Program, Product Management & AI
IIT Patna · 2025 — 2026
Contact details
Contact details
Monika shares this with people who ask. You'll hear back either way.
Ask about this candidate
AI responses may contain errors. Proof status and source information are provided by Proof of Skill.
A Proof CV — one profile, kept current, with the proof attached. Back to the portal