Sachin Keshav

Co-Founder

Noida, Uttar Pradesh, India3 yrs 10 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in developing AI-driven solutions for real-world challenges.
Proven track record in leading technical teams and projects.
Skilled in bridging business objectives with technical execution.

Stackforce AI infers this person is a SaaS-focused AI engineer with strong leadership and technical skills.

Contact

sachin@indusai.app LinkedIn

Skills

Core Skills

Artificial Intelligence (ai)Technical ArchitectureData AnalysisMachine LearningData ScienceSoftware Development

Other Skills

Start-up LeadershipSystem PerformanceSystem ArchitectureLarge Language Models (LLM)Cloud ComputingKubernetesStatisticsWeb TechnologiesPredictionComputer VisionMongoDBREST APIsLLMOpsPyTorchChatGPT

About

A dynamic and passionate engineer, I excel in crafting scalable, production-ready systems with meticulous attention to detail. My enthusiasm for developing AI models is driven by a desire to address real-world industry challenges. Skilled in bridging the gap between business objectives and technical execution, I effectively lead technical teams to transform business concepts into fully-realized AI-driven solutions. My approach combines technical proficiency with collaborative leadership, ensuring successful project outcomes.

Experience

3 yrs 10 mos

Total Experience

1 yr 6 mos

Average Tenure

11 mos

Current Experience

Indus ai

Co-Founder, CTO

Jun 2025 – Present · 11 mos

Architected and deployed low-latency STT → LLM → TTS pipelines for voice agents with sub-500 ms first-response latency, live interruption handling, and multilingual turn-taking.
Led development of custom expressive TTS and STT models, including LoRA-based voice adaptation, GPU-optimized inference, and cost-efficient deployment (₹1–₹3/min economics).
Designed a workflow-driven voice agent platform with stateful memory, intent routing, tool/function calling, and deep integrations with CRMs, databases, calendars, and WhatsApp.
Built and deployed domain-specific voice agents for FMCG ordering, loan recovery, hospitality, healthcare, and enterprise support.
Architected IndusIQ, a real-time natural-language analytics system enabling sub-second insights over large datasets.
Led a 10-member engineering team across ML, backend, and infra; owned systems design, performance tuning, and production reliability.

Start-up LeadershipArtificial Intelligence (AI)Technical Architecture

Emissary

Machine Learning Engineer

Mar 2024 – Jul 2025 · 1 yr 4 mos · Remote

Spearhead end-to-end ML product development, overseeing projects from ideation to production deployment
Architected and implemented an infrastructure platform enabling users to fine-tune and deploy custom Large Language Models (LLMs)
Specialized in LLM fine-tuning, efficient deployments, GPU parallelization, and robust software development
Leveraged LLMs for advanced data extraction techniques, enhancing our data processing capabilities
Orchestrated multi-cloud deployments across AWS and RunPod, ensuring scalability and performance
My role combines cutting-edge ML expertise with practical software engineering, allowing me to deliver innovative AI solutions that drive business value

Large Language Models (LLM)Artificial Intelligence (AI)Machine Learning

Turtlemint

2 roles

Machine Learning Engineer

Jul 2022 – Mar 2024 · 1 yr 8 mos

Built a data extraction service that saved the company’s spending. Automated the system to reduce developer resources, and collaborated with business teams to understand the requirements and continuously upgrade the service and extraction model performance.
Team leading and designing the data flow and architecture for the entire system, ensuring efficient processing and seamless integration of ML models.
Introduced an architecture for ML model monitoring, enabling real-time model performance tracking and automatic training based on MLOps principles.
Implemented NER (Key-Vlaue data extraction) and classification models by developing customized
striding window algorithm, effectively overcoming the 512 token limitation.
Deployed ML models on a Django service and Implemented technologies such as RabbitMQ, Kafka, Slack,
Docker, Kubernetes, and Jenkins to enhance system scalability, efficiency, and automation.
Model development from research phase to production deployment for transformer based models.
LLM based document retrieval and question-answering platform.

KubernetesStatisticsMachine LearningData Science

ML Engineer (Intern)

Jan 2022 – Jun 2022 · 5 mos

Upgraded an existing Django service, created REST APIs to enhance its functionality and performance.
Trained and fine-tuned deep learning models by customising the final layers to achieve accurate multi-
label and multi-class classification results

Web TechnologiesPredictionSoftware Development

Solytics partners pvt. ltd

Data Science Intern

May 2020 – Dec 2021 · 1 yr 7 mos

Developed and implemented a highly efficient Name Matching algorithm, resulting in a significant increase in accuracy and customer acquisition.
Deployed Deep Learning NLP models and core ML models (Fake News classification, text translation, text summarization, semantic search, sentiment analysis) to create a real time news article based information retrieval and News credibility scorer.
Designed variable transformation and data segmentation techniques from user interface to backend-integration,ML Algorithms and Database (PostgreSQL) integration
Time Series Forecasting Algorithms (from EDA, FE, Sampling, Model Estimation and model Testing)
Lead Team of 5 interns to integrate EMR (big data operations using PySpark), S3 bucket, and Django Framework for smooth UI experience

StatisticsWeb TechnologiesData Science