Konstantinos Chervatidis

Zurich, Switzerland · open to remote

Konstantinos Chervatidis

Senior AI/Software Engineer

Senior AI and Software Engineer with 8+ years building production AI products, backend platforms, and financial systems. End-to-end across React and Next.js, Python and Node.js, data infrastructure, cloud deployment, and observability — recent focus on agentic AI and RAG with LangGraph, FastAPI, vector retrieval, and production evaluation workflows.

Available for full-stack and AI roles in Switzerland or remote.

Download CVSchedule a call (opens the booking page in a new tab)Email me

Selected achievements

Measurable outcomes from systems I owned or helped deliver.

Turned an MVP into a repeatable production system

4h → 15min

Built CI/CD pipelines for database migrations, Lambda APIs, backend services, and the React frontend across staging and production. Separately replaced manual AWS provisioning with parameterized Terraform deployments for new trading bots.

Clearstar Labs

Accelerated financial analytics and improved visibility

−70%

Reduced historical-data query latency with Redis and DataLoaders, then built wallet and position tracking with an internal visualization integrated from Figma designs.

Maple Finance

Led delivery of a mobile banking product

4 engineers

Coordinated a cross-functional team and National Bank of Greece stakeholders to unify and classify transactions, support budgeting, and deliver scalable financial-data APIs.

Quintessential SFT

Built monitoring for business-critical financial positions

$5M+

Led four engineers building reliable ingestion, aggregation, retry, monitoring, and alerting workflows, and helped operate the system covering positions across multiple financial venues.

Clearstar Labs

Helped automate infrastructure onboarding

Days → minutes

Owned the internal configuration frontend and contributed to the API, SDK, and service changes that completed the team's automated deployment workflow.

Gelato Network

Selected work

Short demos of products I built end to end — watch before the résumé detail.

Skills

AI & LLM engineering

LLM agentsRAG systemsLangGraphLangChainLangSmithVector searchNeo4jMCP

Languages & frameworks

PythonTypeScriptFastAPINode.jsNestJSReactNext.js

APIs & data

REST APIsGraphQLPostgreSQLRedisSupabaseAsync processingMongoDB

Cloud & infrastructure

AWSAzureGCPTerraformDockerKubernetesCI/CDDeployment automation

Systems & reliability

RabbitMQEvent-driven workflowsTransactional outboxIdempotent jobsInngestMonitoring & alertingGrafanaPrometheus

Experience

  1. Independent AI Engineer

    Jan 2026 – present · Switzerland
    • Built end-to-end AI products spanning React and Next.js interfaces, FastAPI services, retrieval, async workflows, data storage, and cloud deployment.
    • Architected Cortex with stateful LangGraph workflows, citation-backed conversational Q&A, durable ingestion, and end-to-end LangSmith observability.
    • Built an Azure-native RAG system with hybrid semantic retrieval, AG-UI streaming, conversation history, and repeatable Bicep infrastructure.

  2. Software Engineer · Maple Finance

    Sep 2025 – Nov 2025 · Remote
    • Cut primary historical-data endpoint latency approximately 70% with Redis caching and batched data loading.
    • Built an end-to-end wallet and position tracking system, covering real-time indexing, cross-source reconciliation, and an internal visualization delivered from Figma designs with product designers.
    • Contributed to the architecture and implementation of a new lending product, shaping data models and service boundaries.

  3. Senior Backend Engineer · Gelato Network

    Oct 2024 – May 2025 · Switzerland
    • Developed backend services for a high-throughput distributed transaction platform operating across 10+ networks, focusing on reliability, latency, and operational visibility.
    • Owned the internal React configuration frontend and contributed to its supporting APIs and SDK.
    • Optimized a core transaction processing component using a binary search-based approach, meaningfully reducing execution cost and latency.
    • Contributed to complete deployment automation, reducing infrastructure onboarding from days to minutes.

  4. Senior Software Engineer · Clearstar Labs AG

    Sep 2022 – Sep 2024 · Switzerland
    • Took an MVP to production while leading 4 engineers and coordinating requirements with stakeholders and financial analysts.
    • Built end-to-end features to visualize financial data and monitoring for stakeholders, from backend APIs and data models through React dashboards.
    • Built reliable ingestion and aggregation workflows with retries, monitoring, and alerting, supporting visibility into $5M+ in financial positions.
    • Automated delivery across staging and production with CI/CD, cutting releases from hours to approximately 15 minutes, and replacing manual AWS provisioning with parameterized Terraform.

  5. Backend Developer · Quintessential SFT

    Apr 2021 – Aug 2022 · Greece
    • Led a team of 4 engineers on a National Bank of Greece mobile banking project, unifying and classifying transactions and supporting customer budgeting.
    • Owned the backend architecture and coordinated delivery with the bank and project partners, building scalable APIs for high-volume financial data.
    • Improved time-series queries using GraphQL, Redis, and PostgreSQL.

  6. Fullstack Developer · Atypon

    Jan 2020 – Mar 2021 · Greece
    • Built a custom React dashboard over first-party PostgreSQL events and complementary Google Analytics and Tag Manager dashboards.

  7. JavaScript Developer · NMN

    Jun 2018 – Dec 2019 · Greece
    • Developed a browser-based rendering engine powering graphics for more than 20 games.
    • Focused on performant rendering and reliable real-time execution.

  8. Backend Developer · Mita Travel

    Sep 2017 – May 2018 · Greece
    • Implemented high-throughput backend APIs for a booking platform.
    • Worked with a microservices architecture and distributed systems technologies.

Education

MSc in Artificial Intelligence

Aristotle University of Thessaloniki · 2020 – Sep 2022

Thesis: Reinforcement Learning for Forex Trading using Actor-Critic and PPO algorithms

BSc in Computer Science

Hellenic Open University · 2016 – 2020

Thesis: Traffic Sign Classification and Localization with CNNs

Certifications

  • AI Model Development & Deployment

    CourseraJun 2026

  • IBM RAG and Agentic AI

    CourseraMay 2026

  • Machine Learning in Production

    DeepLearning.AIMay 2026

  • Model Context Protocol: Advanced Topics

    Anthropic AcademyMar 2026

  • Building with the Claude API

    Anthropic AcademyMar 2026

Languages

  • English

    Fluent

  • German

    Conversational

  • Greek

    Native

Case studies

Cortex — multi-workflow AI orchestration

View Cortex repository (opens in a new tab)

Problem. Support structured report generation, itinerary planning, and grounded conversational Q&A over user documents — with citations, multi-turn follow-ups, and production-grade ingestion.

Architecture. Stateful LangGraph graphs per workflow; Pinecone as the managed retrieval layer; a transactional-outbox ingestion pipeline drained by Inngest workers; React and Next.js for the product UI; LangSmith tracing across every run.

DocumentsOutboxInngest workersPineconeLangGraph agentsCited answers

Key decisions.

  • Stateful LangGraph architecture over a simpler sequential chain

    Multi-turn follow-ups reuse prior retrieval state instead of re-retrieving each turn, and every response stays grounded in per-run source chunks.

  • Managed Pinecone over self-hosting a vector store

    Retrieval scales as a managed service; operating a vector DB added risk without differentiating the product.

  • Transactional outbox + Inngest workers for ingestion over fire-and-forget async processing

    Idempotent, retry-safe document processing — validated with documents up to 50 pages.

Outcome. A production RAG platform where every answer carries citations to source documents, with observable, replayable ingestion and agent runs.

Azure AI Search RAG — production-shaped Azure-native retrieval

View Azure RAG repository (opens in a new tab)

Problem. Take grounded chat over Markdown and PDF documents beyond a single-script prototype, with citations, streaming, conversation history, corpus controls, and a repeatable security-focused Azure deployment.

Architecture. Azure Blob Storage feeds an Azure AI Search indexing pipeline with chunking and integrated embeddings; FastAPI exposes an Agent Framework AG-UI stream; API Management fronts a private backend; React and Next.js with CopilotKit provide the user interface; Cosmos DB stores per-user discussion history.

Blob StorageAI Search indexerHybrid retrievalFastAPI agentAPIMCopilotKit UI

Key decisions.

  • Hybrid semantic retrieval with integrated vectorization over application-managed embeddings and vector-only search

    Azure AI Search owns index-time and query-time vectorization, then combines lexical and vector matching with semantic reranking for stronger grounding context.

  • AG-UI streaming through Agent Framework over a custom token-streaming protocol

    A standard run and message event model connects the FastAPI agent to the CopilotKit interface while keeping the browser-facing runtime simple.

  • Managed identities and private backend ingress over application API keys and a public backend

    APIM validates the caller, applies quotas, and reaches the internal API using managed identity; service access follows least-privilege RBAC.

Outcome. A tested, documented Azure RAG reference with grounded answers and citations, corpus management, persistent discussions, readiness checks, and repeatable Bicep deployment workflows.