Skip to content
View natiixnt's full-sized avatar
🌴
🌴
17 followers · 5 following

Achievements

Achievement: Pull Sharkx4Achievement: Pair Extraordinairex4Achievement: Public SponsorAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx4Achievement: Pair Extraordinairex4Achievement: Public SponsorAchievement: YOLOAchievement: Quickdraw

Highlights

  • Pro

Block or report natiixnt

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
natiixnt/README.md
Natalia Szczepanik, Senior AI Engineer and Forward Deployed Engineer. LLM agents, RAG, evals, production deployments. Warsaw, Poland, remote or hybrid.

Open to a permanent Senior AI Engineer or Forward Deployed Engineer role. Portfolio | LinkedIn | natjiks@gmail.com

40+ deployments at companies of 20 to 200 people. About 2,770 hours saved per month in total. Cost per task down 43 percent on the agent platform. Legal RAG clause@1 0.86, MRR 0.975. Projects: Agent hub, multi-agent orchestration in production, cost per task down 43 percent, median time down 55 percent, 7 of 38 experiments kept. Legal RAG over 21 Polish statutes, clause@1 0.86, MRR 0.975, 30 of 30 out-of-scope refusals. AAE, self-improving decision system, paper in preparation. Polish embedding model fine-tune, MTEB-PL, in progress.

More: open source and research in the pinned repositories: redcon, flyvis-circuit-bench, n8n-autopilot, crm-autopilot, aws-autopilot, azure-autopilot.

What I do: backend and APIs, custom systems and apps, AI agents and orchestration, RAG, fine-tuning and evaluation, integrations and automation, infrastructure and deployment, security by design.

Stack: LangGraph, LangChain, MCP, Claude, OpenAI, Gemini, pgvector, Qdrant, PyTorch, Python, FastAPI, PostgreSQL, Docker, Kubernetes, AWS, Azure, GitHub Actions, n8n.

Pinned Loading

  1. redcon redcon Public

    Context budgeting layer for coding agents: selects, compresses, caches and budgets repository context. Public benchmark (context-eval), MCP. Open source.

    Python 9 1

  2. chrome-emoji-picker chrome-emoji-picker Public

    Type :name in any Chrome input to search and insert emojis with Tab. Like Discord, but everywhere.

    JavaScript

  3. slayerlabs/slayer slayerlabs/slayer Public

    Protokół dla polskiej inteligencji.

    JavaScript 48 11

  4. flyvis-circuit-bench flyvis-circuit-bench Public

    Benchmark of circuit-discovery methods against exhaustive ablation in a connectome-constrained model of the fly visual system (FlyWire, flyvis). Preregistered, with null models.

    Python