Skip to content

Datt Goswami

End-to-end systems for autonomous coding agents.

Staff-level AI infrastructure engineer — Rust, Python, LLM systems, agent memory, evals, and continual learning.

Calgary, Canada🇨🇦 Canadian citizenOpen to remote

SubstackMediumLinkedInScholarX

Now — Founding/sole engineer at General Intelligence building a synthetic-data marketplace; shipping the Ferrum agent-infra stack in open source alongside.

Selected work

6 projects

  • Featured

    Ferrum Platform

    17+ repos

    Composable platform for autonomous software engineering agents: execution, memory, evals, local inference, gateway, and tooling.

    RustPythonRepoPoster
  • Featured

    cl-agent

    Paper · 6 adapters

    Continual-learning substrate for coding agents — episode capture, replay, skill distillation, sandboxed SOAR search, and LoRA SFT.

    PythonRepoPaper
  • axon

    7× KV compression

    Dynamic batching, SSE streaming, three-tier Metal/CUDA/CPU fallback, TurboQuant KV-cache compression.

    RustRepo
  • secure-agent-runner

    Firecracker-ready

    Policy-checked commands, workspace snapshots, capped output/artifacts, replayable results. Firecracker-ready design.

    RustRepo
  • ferrum-gateway

    HTTP / LLM / MCP

    HTTP, LLM, MCP, and agent traffic with routing, auth, limits, streaming, observability, token budgets, and tool policy.

    RustRepo
  • ap2-rust

    SD-JWT · A2A

    Canonical mandates, SD-JWT/dSD-JWT chains, JOSE crypto, A2A helpers, constraint verification, reference actors, CLI demos.

    RustRepo

Writing

Pinned six

Research

  • Paper · Apr 2026

    cl-agent: A Continual-Learning Substrate for Coding Agents

    Formalizes episode capture, replay, rule-based distillation, and evaluation; baseline-vs.-integrated protocol separates substrate learning from native agent memory.

    Open →
  • Architecture · Apr 2026

    Ferrum Platform Poster

    A composable platform for autonomous SWE: coding agents execute TDD loops, call typed tools, log episodes, replay failures, and measure continual-learning gains across sessions.

    Open →
  • Guide · 2023

    Mastering Prompt Engineering

    Exploratory guide to prompt engineering techniques and patterns for working with large language models.

    Open →
  • AMS · 2017

    Low-Level Wind Climatology Related to Potential Wind Compression

    Reiche, Robinson, Kay, Craun, Goswami, Bass. Wind shear climatology for terminal-radar approach airspace at seven airports.

    Open →

Experience

Founding Software Engineer General Intelligence

Aug 2024 — Present

Calgary, AB

Sole engineer building a two-sided synthetic-data marketplace for AI model training and fine-tuning, from zero to production-deployable.

  • Owned product architecture, system design, implementation, and deployment end to end. Serverless AWS backend (Lambda, S3, DynamoDB, SQS) with a GraphQL API; on-demand scaling with near-zero idle cost.
  • Built smart-contract integrations on SUI Move and Solana (native Rust, no Anchor) for decentralized data provenance and payment settlement. React frontend, Python/Rust/TypeScript backend services.

Senior Software Engineer AdHome Marketing

Sep 2023 — Aug 2024

Calgary, AB

First engineering hire. Scaled the team 1→5, set the technical roadmap, established documentation and code-review practices, ran hiring, and drove ISO 27001 certification engineering changes.

  • Built Sembot, an AI-powered SEO auditing tool backed by a RAG pipeline: Python crawlers (Scrapy) indexed 50+ client sites + competitors into a vector store; GPT-4 grounded audits and DALL-E 3 asset generation replaced weeks of manual review.
  • Developed ProfitFill, an RL-inspired adaptive optimizer reading ServiceTitan technician availability and adjusting Google Ads spend in real time to cut waste when crews were booked and ramp visibility when capacity opened.

Senior Data Integration / Software Engineer InvestCloud

Nov 2021 — Feb 2023

Burlington, ON

Worked in a 6-person pod supporting high-value clients on a B2B wealth-management platform reaching 22M+ end customers, including a $13B AUM firm.

  • Optimized reverse-ETL pipelines with MapDB in-memory caching, removing repeated DB bottlenecks. Built tax-report document generation with tenant-specific ACLs in CouchDB.
  • Operated within a modular monolith of ~1,000 SOAP-based microservices across 5 Java modules; managed monthly Kubernetes deployments via Jenkins; upgraded Plaid API authentication flows.

Senior Software Engineer QuillHash Technologies

Oct 2017 — Oct 2019

India

First engineering hire after the co-founders. Helped recruit, onboard, and technically mentor the team as the company scaled 1→35 engineers across its blockchain security practice.

  • Spearheaded QuillAudits, an automated smart-contract auditing platform that reduced audit turnaround from 2+ months to hours by integrating Mythril and Slither.
  • Automated the end-to-end audit pipeline: GitHub API file ingestion → Mythril/Slither analysis → SQS-coordinated stages → S3 storage → SNS/Lambda client report delivery.

Software Engineer AvMet Applications

Jul 2015 — Oct 2017

Reston, VA

Built weather analysis systems for aviation — accuracy directly affected air-traffic safety.

  • Designed vector-search-based weather pattern matching for the WAVE platform; enhanced NASA's MFCR flight rerouting algorithm.
  • Optimized Python performance 3–5× with Cython; automated NOAA data retrieval (HRRR, GFS, METAR, TAF).

Contact

Email
GitHub
github.com/dattgoswami
LinkedIn
linkedin.com/in/dattgoswami
Substack
datt.substack.com
Medium
medium.com/@dattgoswami
Scholar
Google Scholar
X
@dattgoswami
Where
Calgary, Canada · 🇨🇦 Canadian citizen · Open to remote