---

Senior Software Engineer with 5+ years across the full stack of decentralized systems — from writing and auditing smart contracts to building the off-chain infrastructure that keeps protocols alive in production. Shipped code across DeFi lending markets, AMMs, CDP stablecoins, cross-chain bridges, and NFT ecosystems on both EVM and non-EVM chains.

Beyond Solidity and Move (SUI), I build backend services in Rust and Go — liquidation bots, relayers, keepers, and indexers that handle everything the EVM can't. Lately a lot of that work is agentic: Claude-powered audit pipelines, MCP-driven tool chains, and an automated cross-chain liquidation engine that triggers the moment health factors break.

Open to smart-contract engagements, protocol engineering, security audits, and off-chain infra (relayers, indexers, liquidation bots, keepers). DeFi-focused — lending, perps/AMMs, CDP stables, AA, intents, cross-chain.  Reach out →

---

---

Languages Smart-contract toolchain Web3 libraries Standards (EIPs)

Backend frameworks Chains Cross-chain messaging AI dev tools Infra & monitoring

---

🟢 open · 🟣 merged · 🔴 closed  ·  auto-refreshed twice daily.

 🛠️  Gitlawb / openclaude  ·  runs anywhere. uses anything  ·  99 PRs · 🟣 84 · 🟢 4 · 🔴 11 · all PRs →

• 🟣 #2177 — fix(plugins): anchor marketplace hostPattern against lookalike hosts
• 🟣 #2170 — fix(settings): stop proto-named permission rules from aborting validation
• 🟢 #2169 — fix(api): validate API_TIMEOUT_MS in the native Anthropic clients
• 🟣 #2124 — fix(websearch): reject non-positive WEB_CUSTOM env overrides
• 🟣 #2123 — fix(api): resolve swarm-field tool names by own-property, not the prototype chain
• 🟣 #2100 — fix(repomap): resolve file language by real extension, own-property only
• 🟢 #2099 — fix(provider): match MiniMax and Venice base URLs by hostname, not substring
• 🟣 #2064 — fix(cost): guard model-cost lookup against prototype-member model ids
• 🟣 #2050 — feat(auth): opt-in loopback proxy hosts that keep subscription (OAuth) auth
• 🟣 #2031 — test(user): restore real modules from a pre-mock snapshot
• 🟣 #2028 — fix(sdk): report a permission timeout as a timeout
• 🟢 #2027 — fix(frontmatter): anchor the closing --- to the start of a line
• 🔴 #2024 — fix(api): validate API_TIMEOUT_MS instead of raw parseInt
• 🟣 #2023 — fix(output-style): resolve style names by own-property
• 🟣 #1994 — fix(permissions): anchor the session plan-file match on its exact shape
• 🔴 #1993 — fix(cli): validate --max-turns as a positive integer
• 🟢 #1983 — fix(mcp): resolve server names by own-property, not the prototype chain
• 🟣 #1982 — fix(bridge): truncate derived session titles on grapheme boundaries
• 🟣 #1975 — fix(resume): read the session tag from its own entry, not a tool's tag input
• 🟣 #1974 — fix(memory): match nested directories on path boundaries, not name prefixes
• 🟣 #1966 — fix(commands): insert slash-command argument text literally, not as regex refs
• 🟣 #1965 — fix(cache): use a monotonic clock for conversation-cache LRU recency
• 🟣 #1955 — fix(bash): convert BRE interval braces when previewing sed edits
• 🟣 #1954 — fix(editor): account for NFC boundary composition in insert offset
• 🟣 #1953 — fix(stats): stop over-counting totalDays by one
• 🟣 #1934 — fix(context): order pruned messages by envelope timestamp, not phantom field
• 🟣 #1933 — fix(mcp): preserve ':-' inside ${VAR:-default} default values
• 🟣 #1932 — fix(gitdiff): apply the 1MB diff cap in bytes, not UTF-16 char length
• 🟣 #1918 — fix(memdir): enforce entrypoint cap in bytes, not UTF-16 char length
• 🟣 #1917 — fix(diff): number diff-snippet hunks by their new-file position
• 🟣 #1916 — fix(diff): count dropped lines correctly in truncated diff snippet
• 🟣 #1915 — fix(editor): guard editor-override lookup against prototype keys
• 🟣 #1914 — fix(commands): escape named-argument names before building the regex
• 🟣 #1913 — fix(powershell): make CMDLET_PATH_CONFIG prototype-safe
• 🟣 #1881 — fix(read): report zero lines for an empty file
• 🟣 #1880 — fix(openai-shim): guard tool-arg field lookup against prototype keys
• 🟣 #1873 — fix(diff): don't overcount new-file additions by the trailing newline
• 🟣 #1872 — fix(websearch): match allowed/blocked domains case-insensitively
• 🟣 #1848 — fix(proxy): bypass subdomains for a bare NO_PROXY domain entry
• 🟣 #1847 — fix(permissions): make legacy tool-name alias lookup prototype-safe
• 🟣 #1843 — fix(gitdiff): count in-hunk lines starting with ++/-- in raw diff stats
• 🟣 #1840 — fix(plugins): match reserved-marketplace git URL owner by host, not substring
• 🔴 #1835 — fix(format): show sub-second durations with a decimal instead of "0s"
• 🟣 #1833 — fix(codex): make Codex alias lookup prototype-safe
• 🟣 #1822 — fix(model): resolve [1m]-tagged aliases when 1M context is disabled
• 🟣 #1821 — fix(plugins): keep marketplace reconciliation prototype-safe
• 🟣 #1814 — fix(ollama): cap qwen3-coder-next:cloud output at 32768
• 🟣 #1792 — fix(remote-session): match ingress host by hostname, not raw substring
• 🟣 #1787 — fix(plugins): treat prototype-shadowing marketplace names as not found
• 🟣 #1773 — fix(env-file): collapse escaped backslashes in quoted values
• 🟣 #1760 — fix(bridge): match loopback hostname exactly in HTTPS credential guard
• 🟣 #1710 — fix(code-indexing): guard command detection against prototype-chain names
• 🟣 #1709 — fix(model): preserve [1m] tag for the codex aliases
• 🟣 #1708 — test(install-surfaces): stop execFileNoThrow mock leaking into later suites
• 🟣 #1671 — fix(model): preserve [1m] tag for the 'best' alias
• 🟣 #1669 — fix(provider): match xAI base URL by hostname, not 'x.ai' substring
• 🟣 #1667 — test(file-suggestions): stop cross-spawn mock leaking into later suites
• 🟣 #1652 — fix(worktree): base agent isolation worktree on parent HEAD, not origin/main
• 🔴 #1650 — fix(diff): stop dropping in-hunk lines starting with -- or ++
• 🔴 #1435 — test(attribution): fix order-dependent CI flake in getAttributionTexts
• 🔴 #1434 — fix(update): allow auto-update for OpenClaude builds on third-party providers
• 🟣 #1433 — fix(diff): guard diff language detection against prototype-chain filenames
• 🟣 #1414 — feat(codex-oauth): manual callback URL paste for SSH / remote sessions (#1288)
• 🟣 #1392 — fix(bash): surface rolled-output file path on non-zero exit (#1359)
• 🔴 #1391 — fix(openai-shim): strip image_url blocks on text-only providers (DeepSeek) (#1382)
• 🔴 #1352 — feat(fallback-model): honor --fallback-model in interactive REPL
• 🔴 #1266 — fix(codex-strict): widen optional fields to nullable + strip null from inputs
• 🟣 #1263 — fix(retry): adjust max_tokens on OpenRouter 402 credit shortfall
• 🟣 #1262 — fix(codex-stream): recover tool args delivered only via done events
• 🟣 #1261 — fix(json-schema): support top-level non-object roots via wrap/unwrap
• 🟣 #1260 — feat(safety): warn at startup when 3P provider + permissive mode skip the AI classifier
• 🟣 #1250 — fix(xml): guard escapeXml/escapeXmlAttr against null and undefined (#1247)
• 🟣 #1248 — fix(recovery): keep thinking blocks on resume for reasoning-echo providers (#957)
• 🟣 #1240 — fix(loader): batch markdown reads + cap file size to unblock startup
• 🟣 #1236 — fix(bash): preserve captured stdout in error message on non-zero exit
• 🟣 #1234 — feat(settings): per-model context_window and max_output_tokens overrides
• 🟣 #1178 — feat(provider): add Cloudflare Workers AI integration (#1100)
• 🟣 #1176 — feat(provider): auto-switch on rate limit via providerFallbackChain (#768)
• 🔴 #1169 — fix(query): auto-recover from context-overflow errors
• 🟣 #1164 — feat(model-picker): surface inactive provider profiles in /model (#1119 piece 2)
• 🟣 #1163 — docs(ollama): drop ollama launch openclaude instructions (#1134)
• 🟣 #1070 — fix(openai-shim): redact ?auth=, ?passwd=, ?pwd= in diagnostic URLs
• 🟣 #1068 — perf(local): add OPENCLAUDE_LOCAL_FAST_PATH to skip cloud-only transforms
• 🟣 #1048 — fix(openai-shim): strip store for local providers (vLLM, custom)
• 🟣 #1047 — fix(openai-shim): strip store when baseUrl points at Mistral
• 🟣 #1040 — fix(openai-shim): strip store when baseUrl points at Cerebras (#1023)
• 🟣 #1019 — fix(shims): strip x-anthropic-billing-header from system prompt forwarded to OpenAI/Codex (#607)
• 🟣 #1018 — chore(engines): require Node >=22 to match runtime deps (#1009)
• 🟣 #1006 — fix(web-search): surface diagnostic when adapter returns 0 hits with no native fallback
• 🟣 #959 — fix(openai-shim): strip store when baseUrl points at Gemini
• 🟣 #954 — fix(worktree): surface git stderr in rev-parse failure message (#690)
• 🔴 #953 — fix(model): cap deepseek-v4-pro:cloud / :flash output at 65536 (#917)
• 🟣 #952 — fix(provider): allow remote Ollama without OPENAI_API_KEY (#369)
• 🟣 #864 — fix(startup): url authoritative over model name in banner provider detect
• 🟣 #857 — fix(effort): persist xhigh and send reasoning_effort on chat_completions
• 🟣 #854 — feat(cli): honor --model alone without requiring --provider
• 🟣 #840 — fix(test): add missing teammate exports to hookChains integration mock
• 🟣 #837 — feat: add OPENCLAUDE_DISABLE_TOOL_REMINDERS env var to suppress hidden tool-output reminders
• 🟣 #826 — docs(env): document OPENCLAUDE_DISABLE_STRICT_TOOLS in .env.example

 🛠️  Pablosinyores / aether  ·  Low latency arbitrage flash loan mechanism  ·  73 PRs · 🟣 70 · 🔴 3 · all PRs →

• 🟣 #189 — release: backrun reject observability — info logs + deficit histograms
• 🟣 #188 — obs(metrics): histograms for backrun gross / gas / deficit per reason
• 🟣 #187 — obs(sim): surface backrun reject reasons in stdout by default
• 🟣 #183 — perf(sim): batch V2 reserves prewarm via Multicall3 (~95% CU saving on storage path)
• 🟣 #182 — debug(sim): surface victim + arb revert reasons instead of dropping output
• 🟣 #181 — obs(grafana): cache + RPC-reduction dashboard
• 🟣 #180 — perf(sim): write fresh V2 reserves back into cache after RPC fetch
• 🟣 #179 — perf(metrics): expose pre-warm cache hit / RPC-fetch counters
• 🟣 #178 — perf(sim): recover bytecode cache + V2 reserves cache wiring
• 🟣 #177 — docs(readme): refresh for mempool backrun, decoder coverage, cache tier
• 🟣 #176 — perf(sim): WS-fed V2 reserves cache, skip getStorageAt for warm pools
• 🟣 #175 — perf(sim): wire bytecode cache into prewarm + engine + mempool pipeline
• 🟣 #174 — perf(sim): persistent on-disk bytecode cache (redb)
• 🟣 #173 — perf(sim): cap pre-warm RPC concurrency at 8 in-flight
• 🟣 #172 — fix(decoder): u64 overflow in bancor test minReturnAmount
• 🟣 #171 — feat(decoder): Universal Router execute() + V3/V2 opcodes + V4_SWAP named-skip + real-mainnet fixtures
• 🟣 #170 — feat(decoder): silent-skip 1inch v6 limit-order admin selectors
• 🟣 #169 — fix(decoder): use real Solidity names for UniV3 SwapRouter02 overloads
• 🟣 #168 — feat(decoder): silent-skip UniV2 liquidity-management selectors
• 🟣 #167 — fix(demo): align shutdown SQL columns with current schema
• 🟣 #166 — feat(demo): 1-day shadow-mode end-to-end demo script
• 🟣 #165 — feat(simulator): inject AetherExecutor bytecode into revm CacheDB for shadow runs
• 🟣 #164 — feat(mempool): decode 1inch v6 AggregationRouter swaps
• 🟣 #163 — feat(mempool): decode UniV3 multicall(bytes[]) for nested swaps
• 🟣 #162 — feat(mempool): Bancor V3 multi-hop post-state prediction (two-pool path)
• 🟣 #161 — feat(simulator): revm reader for Curve + Balancer post-state replay
• 🟣 #160 — feat(mempool): route Bancor V3 through analytical post-state predictor
• 🟣 #159 — feat(mempool): route Curve through analytical post-state predictor + filter pool addresses
• 🟣 #158 — feat(mempool): revm post-state replay fallback for V3 tick-crossing swaps
• 🟣 #157 — feat(decoder): Bancor V3 BancorNetwork trade() decoder + protocol tagging
• 🟣 #156 — feat(decoder): Curve pool-direct exchange() decoder + protocol tagging
• 🔴 #154 — feat(metrics): mempool first-seen → inclusion latency histogram
• 🟣 #147 — feat(mempool-backrun): Phase 1 Part C — risk gates + shadow rollout + live validation
• 🟣 #143 — feat(mempool-backrun): Phase 1 Part B — Go executor mempool bundle builder + fan-out
• 🟣 #142 — feat(mempool-backrun): Phase 1 Part A — Rust validate_backrun revm sim + gRPC publish
• 🟣 #141 — docs(research): tx-ordering strategy + builder matrix + Phase 1 staged issue bodies
• 🟣 #130 — feat(engine): multi-signal candidate gating for detected arb cycles
• 🟣 #129 — fix(simulator): rewrite wss/ws RPC URL to https/http for fork backend
• 🟣 #128 — feat(pools,engine): V3 / Curve / Balancer post-state predictors + state cache
• 🔴 #127 — feat(pools,engine): V3 / Curve / Balancer post-state predictors + state cache
• 🟣 #122 — feat(ledger): Go PgLedger pgxpool impl + executor wiring (PR-2 of #95)
• 🟣 #119 — feat(ledger): PgLedger sqlx impl + engine wiring + bounded writer (PR-1 of #95)
• 🟣 #118 — feat(mempool): live tracking — subscribe + decode + analytical V2 post-state sim
• 🟣 #116 — chore(observability): PR #83 follow-ups (webhook hygiene, alertmanager watchdog, cold-start gating + 6 nits)
• 🟣 #115 — feat(ledger): trade-ledger foundation — schema + no-op access layer
• 🟣 #114 — fix(provider): adaptive HTTP polling for local/Anvil endpoints
• 🟣 #109 — feat(metrics): split aether_decode_errors_total by reason
• 🟣 #108 — docs: add mermaid architecture diagrams
• 🟣 #105 — feat(replay): production-parity optimizer + tipBps in aether-replay
• 🟣 #104 — fix(replay): deploy AetherExecutor to Anvil fork in e2e script
• 🟣 #103 — chore(observability): land PR #84 content on main (cherry-pick recovery)
• 🔴 #102 — chore(observability): land PR #84 content on main (post-stack recovery)
• 🟣 #94 — fix(ingestion): decode Uniswap V2 / Sushi Swap events instead of dropping
• 🟣 #93 — feat(shadow): live mainnet shadow-mode runner + WS transport preference
• 🟣 #88 — feat(replay): aether-replay binary — historical block + intra-block MEV detection
• 🟣 #87 — chore(clippy): fix 12 workspace lint errors (#82)
• 🟣 #85 — feat(grpc-server): surface decoder drops via warn log + counter
• 🟣 #81 — fix(executor): replace production stubs with on-chain validation (#66)
• 🟣 #79 — perf(build): tune [profile.release] with fat LTO and single codegen unit
• 🟣 #64 — test(infra): end-to-end staging validation of full pipeline
• 🟣 #62 — perf(detector): parallel revm simulation with CacheDB pre-warming
• 🟣 #61 — perf(grpc): switch to UDS transport and reduce serialization overhead
• 🟣 #60 — feat(metrics): instrument remaining observability spec metrics
• 🟣 #57 — chore(infra): remove dead postgres/redis dependencies
• 🟣 #56 — feat(monitor): add Prometheus metrics to Rust engine and Go executor
• 🟣 #52 — fix(executor): wire gas oracle and nonce manager to real RPC
• 🟣 #49 — fix(docker): wire env, observability, and local node for Docker stack
• 🟣 #46 — fix(build): align Rust and Go versions across Docker, Ansible, docs, and gRPC config
• 🟣 #43 — feat(executor): dynamic tip share based on inclusion rate
• 🟣 #42 — fix(risk): tune revert circuit breaker for flash loan execution
• 🟣 #41 — fix(executor): implement transaction signing with go-ethereum
• 🟣 #38 — fix(engine): bootstrap pools from pools.toml at startup
• 🟣 #15 — chore: add dev setup files and env configuration

 🛠️  meilisearch / meilisearch  ·  A lightning-fast search engine API bringing AI-powered hybrid search to your sites and applications.  ·  9 PRs · 🟣 1 · 🟢 2 · 🔴 6 · all PRs →

• 🟢 #6496 — Respect pagination.maxTotalHits in total hits metadata
• 🟢 #6483 — Support target index names in the export route
• 🟣 #6468 — Split unit tests into separate files (one commit per file)
• 🔴 #6465 — Split facet range search unit tests into a separate file
• 🔴 #6464 — Split facet incremental unit tests into a separate file
• 🔴 #6461 — Split matches unit tests into a separate file
• 🔴 #6460 — Split permissive-json-pointer unit tests into a separate file
• 🔴 #6453 — Split index_documents unit tests into a separate file
• 🔴 #6452 — Report word count while building the words FST

 🛠️  lightninglabs / lightning-terminal  ·  Lightning Terminal: Your Home for Lightning Liquidity  ·  8 PRs · 🟣 4 · 🟢 3 · 🔴 1 · all PRs →

• 🟢 #1341 — test: add coverage for normalizeVerString
• 🟣 #1337 — config: make litd -V report litd version
• 🟢 #1335 — litcli: drop unused flags from addinvoice command
• 🟣 #1334 — accounts: add test coverage for checkLabel
• 🟣 #1311 — app: make build/.gitkeep postbuild step git-worktree friendly
• 🟢 #1299 — dev.Dockerfile: allow forcing faraday repo and version
• 🔴 #1296 — litcli: align help text style for top-level commands
• 🟣 #1295 — docs: note that FreeBSD users need gmake to build

 🛠️  oven-sh / bun  ·  Incredibly fast JavaScript runtime, bundler, test runner, and package manager – all in one  ·  5 PRs · 🟢 4 · 🔴 1 · all PRs →

• 🟢 #32638 — fix(console): send console.trace to stderr with a "Trace:" prefix
• 🟢 #32482 — fix(worker): don't keep the main thread alive for a global message listener
• 🟢 #32481 — fix(node:perf_hooks): report nodeTiming milestones as offsets from timeOrigin
• 🟢 #32460 — fix(node:util): correct numericSeparator for negative fractional numbers
• 🔴 #32459 — feat(node:buffer): implement transcode()

 🛠️  solana-playground / solana-playground  ·  Quickly develop, deploy and test Solana programs from browsers  ·  5 PRs · 🟣 3 · 🟢 2 · all PRs →

• 🟣 #472 — Suggest solutions for common build errors
• 🟢 #471 — Fix enum variant case conversion not matching Anchor
• 🟢 #470 — Display Rust Analyzer status in the bottom bar
• 🟣 #466 — Automatically fetch on-chain IDL when a program is deployed
• 🟣 #465 — Add ability to scroll past end in test UI

 🛠️  Infisical / infisical  ·  Infisical is the open-source platform for secrets, certificates, and privileged access management.  ·  4 PRs · 🟢 4 · all PRs →

• 🟢 #6927 — fix(keystore): use strconv instead of rune conversion in memory IncrementBy
• 🟢 #6902 — fix(project): allow spaces in All Projects search query
• 🟢 #6901 — fix(membership): correct inflated total count in user and identity list queries
• 🟢 #6884 — fix(secrets): remove personal overrides when moving secrets

 🛠️  PlakarKorp / plakar  ·  plakar is a backup solution powered by Kloset and ptar  ·  3 PRs · 🟣 2 · 🟢 1 · all PRs →

• 🟣 #2301 — ptar: allow filtering the snapshots pulled from a kloset
• 🟢 #2295 — ls: report the errors recorded in a snapshot
• 🟣 #2294 — mount: set a valid modification time on the root directory

 🛠️  estuary / flow  ·  🌊 Continuously synchronize the systems where your data lives, to the systems where you _want_ it to live, by managing your data flows with Estuary. 🌊  ·  3 PRs · 🟢 1 · 🔴 2 · all PRs →

• 🔴 #3184 — docs: per-command Docker requirements for flowctl (#2968)
• 🔴 #3173 — notifications: use "Estuary" in alert email subjects (#3000)
• 🟢 #3172 — fix(flowctl): honor -q short flag on raw get (#3084)

 🛠️  lightninglabs / lndinit  ·  cloud infra tooling for lnd provisioning/unlocking  ·  3 PRs · 🟢 3 · all PRs →

• 🟢 #86 — migrate-db: warn loudly before writing irreversible tombstones
• 🟢 #85 — migrate-db: skip missing source DBs by default
• 🟢 #84 — migration: default to sqlite and surface flag defaults in -h

 🛠️  Syndica / sig  ·  a Solana validator client implementation written in Zig  ·  2 PRs · 🟢 2 · all PRs →

• 🟢 #1786 — feat(init): back large shared memory regions with huge pages
• 🟢 #1785 — feat(shred-receiver): add prometheus metrics and grafana dashboard

 🛠️  future-agi / future-agi  ·  Open-source, end-to-end platform for evaluating, observing, and improving LLM and AI agent applications. Tracing · Evals · Simulations · Datasets · Gateway · Guardrails. Self-hostable. Apache 2.0.  ·  2 PRs · 🟢 2 · all PRs →

• 🟢 #2360 — feat(observe): add a free-text search box to the traces toolbar
• 🟢 #2211 — fix: guard EE usage metering fallbacks so OSS dataset creation does not crash

 🛠️  juspay / hyperswitch  ·  Open source, composable payments platform \| PCI compliant \| SaaS and Self-host options \| Enables connectivity to multiple payment, payout, fraud, vault and tokenization providers \| Uplifts authorization with intelligent routing and revenue recovery \| Reduce payment processing costs with cost observability \| Reduces payment ops with reconciliation  ·  2 PRs · 🟢 2 · all PRs →

• 🟢 #13495 — test(currency_conversion): assert conversion results and fix swapped rate fixtures
• 🟢 #13494 — fix(events): map PaymentPartiallyAuthorized to the payments event class

 🛠️  lightningnetwork / lnd  ·  Lightning Network Daemon ⚡️  ·  2 PRs · 🟢 2 · all PRs →

• 🟢 #10989 — lnrpc+rpcserver: include short channel id in pending close channels
• 🟢 #10784 — cmd/commands: read full stdin for lncli unlock --stdin

 🛠️  uutils / coreutils  ·  Cross-platform Rust rewrite of the GNU coreutils  ·  2 PRs · 🟣 2 · all PRs →

• 🟣 #13640 — shred: report the real error for an inaccessible file
• 🟣 #13637 — chmod: report the real error when a target's metadata is inaccessible

 🛠️  ag2ai / ag2  ·  AG2 (formerly AutoGen): The Open-Source AgentOS.Join us at: https://discord.gg/sNGSwQME3x  ·  1 PR · 🟣 1 · all PRs →

• 🟣 #2789 — Fix LLMConfig for 5 notebooks (7)

 🛠️  anza-xyz / agave  ·  Web-Scale Blockchain for fast, secure, scalable, decentralized apps and marketplaces.  ·  1 PR · 🟢 1 · all PRs →

• 🟢 #14185 — cli-config: don't panic computing websocket url for port 65535

 🛠️  lightninglabs / loop  ·  Lightning Loop: A Non-Custodial Off/On Chain Bridge  ·  1 PR · 🟣 1 · all PRs →

• 🟣 #1133 — staticaddr: move address.Parameters to script package

 🛠️  openobserve / openobserve  ·  Open source observability platform for logs, metrics, traces, frontend monitoring, pipelines and LLM observability. A sophisticated, simple and highly performant alternative to Datadog, Splunk, and Elasticsearch with 140x lower storage costs and single binary deployment.  ·  1 PR · 🟢 1 · all PRs →

• 🟢 #13806 — fix: reject empty org_id in POST /api/{org_id}/users

---

Pinned on my profile  ·  auto-refreshed twice daily.

Project	Stack	What it does
chainscope	Rust PLpgSQL	A Uniswap V3 indexer on Ethereum mainnet, written in Rust. Exactly-once writes, reorg-safe, resumable.
phantom-filler	Solidity Rust	A high-performance, Rust-based intent-based order execution engine inspired by how UniswapX fillers operate. The system monitors mempools and order books across multiple EVM chains, discovers signed swap intents (Dutch auction orders), evaluates optimal fill strategies using on-chain and off-chain liquidity, and executes fills to capture profit
Meridian	Solidity Go	Non-custodial digital-asset prime brokerage: ERC-4626 lending pools, constrained on-chain margin accounts with up to ~5x leverage through whitelisted adapters, an off-chain portfolio margin engine, and liquidation keepers.
chainpulse	Go TypeScript	Multi-chain blockchain event indexer with AI agent query layer (Go + Kafka + ClickHouse + MCP).
relay	Rust PLpgSQL	Outbound webhook delivery service in Rust — signed, queued, retried, with circuit breaking and per-endpoint rate limiting.
crossval-plan-tracker	TypeScript JavaScript	—

---

I take on Solidity audits, threat modeling, and invariant-suite design for protocol launches. Workflow is Slither + Foundry invariants + manual review against an explicit threat model — augmented by an in-house Claude-powered audit pipeline that runs reentrancy / access-control / oracle / arithmetic checks before manual passes. Reach out via LinkedIn for scope and engagement details.

• Security-Audit-Agent — internal multi-agent automation I use to bootstrap reviews.

---