Ashish Adhikari

Ashish Adhikari.

Senior Data & AI Engineer.

Work Experience

Founder & Engineering Lead

2026 - Present

YoursSherpa Remote

  • Founded a forward-deployed engineering practice that builds data platforms and AI agents inside client teams, in the client's own cloud.
  • Set the delivery approach: deterministic code where the rules are clear, AI agents only where work needs reading or judgment, and human approval for anything that sends, pays or commits.
  • Built eight reference architectures: RAG and agentic RAG, data engineering automation with a private LLM, workflow agents, MCP servers and skills, warehouses, lakehouses, AI harness and fine-tuning, and analytical dashboards.

Senior Software Engineer

Jul 2025 - Present

Abacus Insights Boston, USA (Remote)

  • Designed and maintained large-scale batch data pipelines in Databricks on AWS to standardize US healthcare datasets.
  • Implemented medallion architecture (Bronze/Silver/Gold) with Delta Lake, including SCD handling.
  • Orchestrated jobs using AWS Glue and Airflow; leveraged EventBridge and Lambda.
  • Published curated tables to Snowflake for downstream analytics.

Senior Data Engineer

Oct 2024 - Jun 2025

Cedar Gate Technologies Greenwich, CT, USA (Remote)

  • Developed an AWS Redshift data warehouse storing 10M+ healthcare records annually.
  • Improved query performance by 30% and reduced reporting delays by 70%.
  • Automated pipelines with AWS Glue, Airflow, and EventBridge-driven schedules.

Data Engineer

Jan 2023 - Oct 2024

Cedar Gate Technologies Greenwich, CT, USA (Remote)

  • Engineered robust ETL pipelines processing 500K+ healthcare claims daily.
  • Designed data validation frameworks, reducing data quality issues by 85%.

Associate Data Engineer

Jan 2022 - Dec 2022

Cedar Gate Technologies Greenwich, CT, USA (Remote)

  • Assisted in developing data processing workflows for healthcare datasets.
  • Contributed to SQL optimization efforts and database performance tuning.

Remote AI/ML Specialist

Oct 2021 - Jan 2022

Clevero Melbourne

  • Assessed ML implementation for applications, reducing decision lead time by 40%.

Certified Engineer

Industry-recognised professional certification

Databricks Certified Data Engineer Associate
Professional Certification

Databricks Certified Data Engineer Associate

Databricks  ·  2024

Validates proficiency in building, managing, and optimizing data pipelines on the Databricks Lakehouse Platform. Covers Apache Spark, Delta Lake, Delta Live Tables, Databricks Workflows, and Unity Catalog governance.

Apache SparkDelta LakeDelta Live TablesDatabricks WorkflowsUnity CatalogPython / SQL
Verify Credential

YoursSherpa

My forward-deployed engineering practice for data and AI

YoursSherpa

Founder & Engineering Lead

2026 – Present

Runs on

DatabricksAWSAzureGoogle CloudSnowflakePrivate cloud
Visit yourssherpa.com

Data platforms and AI agents

Code where the rules are clear. Agents where they aren't.

YoursSherpa works inside client teams to map the manual work and build pipelines, retrieval systems and AI agents in the client's own cloud. Each step is automated with deterministic code where it can be, given to an AI agent only where the work needs reading or judgment, and approved by a person when it sends, pays or commits.

Eight reference systems

  • 01RAG and agentic RAGAnswers from your documents with citations, access control and evaluation.
  • 02Data engineering automationIngestion, Silver mappings and validation, assisted by an LLM hosted inside your network.
  • 03Workflow automation and AI agentsManual processes automated in code, with agents only where judgment is needed.
  • 04MCP servers and skillsSkills, data and content served to any MCP-compatible AI tool from one governed endpoint.
  • 05Data warehousingStar-schema warehouses tuned for the reports teams run every day.
  • 06Data lake and lakehouseMedallion lakehouses in open formats, with quality gates and history.
  • 07AI harness and fine-tuningInstructions, tools, guardrails and evaluations first. Fine-tuning when evaluations call for it.
  • 08Analytical dashboardsMetrics defined once, reconciled against source, in the BI tool you already use.

How an engagement runs

  1. 01AssessMap the process as it really runs, workarounds included.
  2. 02FeasibilityCode, agent or person for each step, with a written go or no-go.
  3. 03BuildIn the client's environment, starting with the smallest piece that proves it.
  4. 04Run and hand overMonitoring, evaluation, runbooks and training.

Honors & Awards

Abacus Hackathon Winner 2025
Abacus Hackathon Winner 2025
ICT Meetup V6.0 Winner
ICT Meetup V6.0 Winner
Imagine Cup Winner 2021
Imagine Cup Winner 2021
LOCUS 2020 Winner
LOCUS 2020 Winner

Other Certifications

Continuous learning across AI, data, and cloud technologies

Google Foundations: Data, Data, Everywhere
Google Foundations: Data, Data, Everywhere
Ask Questions to Make Data-Driven Decisions
Ask Questions to Make Data-Driven Decisions
LinkedIn: Agentic AI for Developers
LinkedIn: Agentic AI for Developers
Agentic AI for Developers – Concepts & Application
Agentic AI for Developers – Concepts & Application
IBM Python for Data Science
IBM Python for Data Science
Neural Networks & Deep Learning
Neural Networks & Deep Learning
Improving Deep Neural Networks: Hyperparameter Tuning
Improving Deep Neural Networks: Hyperparameter Tuning
Structuring Machine Learning Projects
Structuring Machine Learning Projects
Machine Learning – Zero to Hero
Machine Learning – Zero to Hero
Introduction to Large Language Models
Introduction to Large Language Models
What Is Generative AI?
What Is Generative AI?
Transfer Learning for NLP with TensorFlow Hub
Transfer Learning for NLP with TensorFlow Hub
Apache Spark – Beyond Basics
Apache Spark – Beyond Basics
HackerRank SQL
HackerRank SQL
Google Foundation Data Certificate
Google Foundation Data Certificate