Pavan Modi

+•• ••••• •••••••••••@•••••.com

5.1 yrs

As a Machine Learning Engineer with 3 years of experience, I am driven by a determined pursuit of mastery, constantly seeking autonomous challenges where I can apply logical, pragmatic solutions. I thrive on accountability and am ambitious in delivering impactful results, approaching complex problems with a focused yet relaxed demeanor.

  • Remote

Ask about Pavan

Skills

Claimed

MLflow UsageVLLM UsageDecision MakingMentoring and Peer CoachingData ExaminationConceptual UnderstandingPortfolio ManagementWealth ManagementProblem SolvingAdaptabilityBusiness AcumenStakeholder Management

Proof of Experience

Machine Learning Engineer

Fuzzy Labs · Feb 2024 – Mar 2026

  • Co-led a RAG system over sensitive UK government legal documents.
  • Used Qdrant as vectorDB and moved retrieval from semantic-only to hybrid (BM25 + dense) with reranking, fixing exact statute/clause lookups cosine search missed; measured the gain on NDCG@10 and MRR against a 660-query eval set built with government lawyers.
  • Served it with vLLM behind an API-gateway auth layer that rejected unauthenticated queries before any LLM spend, added semantic caching to serve repeated queries without re-hitting the LLM, with Langfuse tracing, cutting query latency ~16%.
  • Fine-tuned a pre-trained LLM (Mistral) on domain-specific UK legal data using QLORA (4-bit) to adapt it efficiently for training and production serving; evaluated the fine-tuned model against the RAG baseline and recommended against shipping after inference cost and latency did not justify the marginal business gain.
  • For a real-estate internal-document chat product, designed and shipped a graph RAG system (PostgreSQL with pgvector) to replace underperforming hybrid RAG after diagnosing the data as inherently relational.
  • Owned the evaluation, built the eval set, implemented hard-negative mining, and wrote LLM-as-judge scoring on context and answer relevance, run through the team's eval-gated CI before release.
  • Managed conversation context with query rewriting, HyDE, and reranked chunk shortlisting, plus a sliding-window memory that I extended with LLM summarization once the window alone lost earlier context.
  • Explored inference-optimization techniques, quantized an LLM to GGUF (llama.cpp) for on-device mobile deployment in a POC, evaluating the cost and privacy trade-offs of moving generation to the edge.
  • For an e-bike rental platform, built an event-driven pipeline with Kafka to stream and process a steady flow of rental, return, and availability events in near real time, with services communicating over gRPC and REST, Dockerised and deployed on AWS behind a load balancer, with blue-green deployment and Prometheus/Grafana monitoring.

Machine Learning Engineer

Barclays · Nov 2020 – Nov 2022

  • Owned the full ML pipeline: Docker, MLflow for tracking, CI/CD with automated retraining and rollback.
  • A/B tested new model versions against production baselines before rollout.
  • Cut deployment time from ~2 weeks to under 10 days.
  • Improved existing fraud detection models in a regulated banking environment using LightGBM and feature engineering over large transactional datasets (SQL), tuned for card fraud and suspicious transactions across 2.8M+ users, which reduced false positives by 12%.
  • Tried a deep learning approach but latency and cost did not justify the marginal gain so stuck with what worked.
  • Set up monitoring with Prometheus, Grafana and evidently ai for latency, drift, model performance dashboards.
  • Caught a distribution shift early that would have degraded fraud recall by ~8%.
  • Kept SLAs intact, saved compute costs by 15%.
  • Worked across risk, compliance, and product teams on loss-prevention workflows.
  • Every model decision needed documentation, audit trails, and explainability reports, not optional in banking.

Data Scientist

Digi-Soch · Oct 2019 – Oct 2020

  • Built ETL pipelines in SQL and time-series forecasting models (ARIMA, Prophet) for retail and e-commerce clients.
  • The demand forecasting work fed directly into inventory planning, one client reduced overstock by ~18% (in Q1) after deployment.

Proof-of-Work

No completed sessions yet. Anything Pavan finishes shows up here with the evidence behind it.

Proof of Education

MSc, Data Science

University of Greenwich, London · 2023 — 2024

A Proof CV — one profile, kept current, with the proof attached. Back to the portal